Abstract
The miniature high-speed digital valve is essentially a kind of miniature system that contains many nonlinear factors and involves multifield coupling among electromagnetic, mechanical, and liquid ones. In this study, the dimension of an electromagnetic structure is normalized into a unified variable, and the numerical analysis method is used to optimize the values of each parameter size on the premise of satisfying the requirements. The electronic control board is the control signal input equipment of the valve. The input changes of voltage and current directly affect the thermal characteristics of the valve. The electronic control board is the power source of the digital valves, and their high speed partially depends on the characteristics of the electronic control board. In this study, a control board for miniature digital valves was developed, which realizes the rapid rise and fall of current and meets the design requirements. A new electronic control board is manufactured, and a control platform for the electronic control board is built. Experiments demonstrate that there is no delay in the current rise and fall, showing that the valve achieves its dynamic response target. Using the finite element analysis software Workbench to simulate it numerically, the temperature field distribution in the original structure is not uniform, which affects the magnetic properties of the electromagnetic field. To address this disadvantage, this study proposes to insulate the heat source with thermal insulation material, which will not affect the temperature distribution of the entire electromagnetic components. The simulations show that the proposed method is effective and reliable.
1. Introduction
Different from the proportional and servo technologies, the digital hydraulic system adopts a miniature high-speed digital valve with switching characteristics. The advantages of the miniature high-speed digital valves include their low cost, high reliability, and antipollution properties. Because the digital hydraulic system usually has the structural characteristics of multiple valves in parallel, the valves used in digital hydraulic systems should be miniature, high-speed, and digital valves.
Lonnie et al. proposed a miniature-hydraulic valve, that uses memory alloy material [1]. The excited voltage can be used to induce miniature-deformation and expansion of the memory alloy material in the direction of the valve’s length, hence driving the valve’s opening. The dynamic response of the valve can reach 200 Hz. It can control a humanoid manipulator and is suitable for use in special medical devices. The flow rate of the memory alloy valve is only 10 mL/min at a pressure drop of 20.7 MPa. Hydraulic components are usually measured in liters, but this small valve is measured in milliliters. It is true that traditional flow control may be archived using parallel valve technology, but many small valves need to be connected in parallel. This may not be fair in terms of the economy and reliability. Zhang et al. developed a miniature solenoid valve that can be applied to an intelligent switch slipper and is suited for oil and gas well that operate at high temperature and high pressure [2]. Considering the special operating conditions, the miniature valve is 18 mm in diameter and has a maximum working pressure of 21 MPa and a flow rate of 2 L/min at a pressure drop of 3.5 MPa and a temperature of 25°C. Yang et al. conducted a thorough investigation into the miniature high-speed digital valve, measuring its electromagnetic field, electromagnetic force, and flow characteristics [3–5].
Karvonen et al. developed the first generation of miniature solenoid valves with an outside diameter of 10 mm and an opening and closing time of 1.9 ms and 2.2 ms, respectively. Due to the use of stainless steel with inadequate magnetic insulation, the magnetic circuit experiences a “circuit break” [6]. Puumala et al. developed the second generation of miniature solenoid valves. The outside diameter of the miniature valves in this generation has been increased to 11 mm [7]. The opening energy consumption of the valves was decreased to a minimum of 57 mJ. The open response time is 2.8 ms, whereas the close response time is 2.2 ms. Paloniitty et al. developed the third generation of miniature solenoid valves [8]. The outside diameter of this generation of miniature valves was redesigned to 10 mm. The opening response time is 2.0 ms, and the closing response time is 2.8 ms when the test pressure is 21 MPa. The time required for the valves to close greatly varies among individuals. Linjama et al. developed the fourth-generation miniature solenoid valve with an exterior diameter of 10 mm. Under various pressure drops, the opening delay time ranged from 1.4∼2.3 ms, and the closing delay time was between 2.0 ms and 3.4 ms [9]. The limitation of the fourth-generation miniature-valve is an increase in working pressure, which is currently at 21 MPa.
The initial power of valve core movement is voltage [10, 11]. In comparison to other factors, such as hydraulic pressure, inertia, and friction, voltage is the decisive factor. Many scholars have studied a variety of drive boards to pursue the high-speed digital valve [12–14]. Lee developed a three-power drive board for the high-speed on-off valves. The traditional driving circuit of the valve controls the Darlington tube by optical coupler, and the driving voltage is constant [15]. Guo et al. Considered that low driving voltage cannot satisfy the requirement for fast pull-up current, a high-low voltage dual-power drive module was designed [16]. The optimized driving circuit shortened the current pull-up time to 105 μs and a closing delay time to 1.05 ms. Song et al. developed a double-winding high- by adding a quick-start coil to the standard high-speed solenoid valve used in liquid rocket engines [17]. A chip integrated control circuit was built in the valve. The voltage switching mode of high-voltage starting and low-voltage holding was realized by optimizing the time of the two coils.
Guo et al. proposed an improved boost circuit based on complex programmable logic device (CPLD) control logic [18]. The circuit adopted step-by-step comparison of boost mode and pulse width modulation according to CPLD logic to achieve the goal of fast opening and low energy consumption. The increase time of current is 150 μs, and the recovery time of driving voltage is 200 μs. Zhao et al. analyzed the effects of different driving voltages on the dynamic response and energy loss of high-speed solenoid valves by finite element method [19]. The relationship among excitation voltage, dynamic response, and energy loss was established by energy transfer theory. The excitation voltage was optimized to maximize the energy efficiency of the high-. Cheng et al. studied the energy loss and dynamic response of an ultra-high-speed electromagnetic injector through different driving strategies [20]. The transient electromagnetic finite element method was used to calculate the temperature field, dynamic response, and energy loss of an ultra-high-speed electromagnetic injector.
The characteristics of magnetic field determine the rising speed and final value of electromagnetic force, and the primary factors affecting magnetic field include magnetic field temperature, types of magnetic materials, permeability, and magnetic saturation intensity. Temperature has an effect on the magnet’s performance, depending on the properties of the selected material. To stabilize the magnetic field characteristics, the electromagnetic components temperature should be kept constant. Previous research on solenoid valves focused on mechanical, electromagnetic, and flow fields but ignored the effect of the heart source field on temperature rise of electromagnetic coil [21, 22].
Newton’s formula and the thermal circuit method were used to account for the effects of heat radiation, convection, and conduction on the temperature rise of solenoid valves. However, due to the fact that distinct heat problems have different boundary conditions, these calculating methods mostly rely on empirical data. As a result, the calculation results are suboptimal. Another method for calculating the temperature rise of electromagnet is based on heat transfer and hydrodynamics theory. By numerical calculation, the temperature distribution of the magnetic field is obtained. At the moment, the finite element method is widely used [23, 24]. The power loss in electromagnetic analysis was used as the thermal load in thermodynamic simulation using the ANSYS Workbench simulation software platform [25, 26].
Many research works on the miniaturization, digitization, and thermal analysis of high-s have been published in the aforementioned literature. According to these papers, it is critical for high-s to minimize the geometric size of the structure, digitally control the signal, and conduct thermal analysis. The aim of this paper is to study the minimization of structural parameters, signal digitization, and thermal analysis. As a result, the structural minimization method and related test results completed the verification of the theory.
2. Miniaturization
The performance of the high-speed digital valves depends on the appropriateness of the electromagnetic structure in a great measure. Because the size of the electromagnetic structure of the digital valve limits the size of the electromagnetic force, which drives the movement of the valve core and restricts the movement acceleration and speed, the size of the structure limits the performance of the digital valve. This paper studies a DC solenoid electromagnet with a simple and reliable structure. Its structure sketch is shown in Figure 1.
For flat head electromagnets, the electromagnetic force expressed as a function of geometric size is determined by the following formula [27]:where is electromagnetic force, is heat dissipation coefficient of coil outer surface, is coil rising temperature, is filling coefficient of coil, is coil outer diameter, is core radius, is difference coefficient of heat dissipation condition between inner and outer surface of coil, is permeability coefficient, is calculated cross-section area of the working air gap, is electrical conductivity of the wire resistance coefficient, is specific leakage permeability, is parameter, , is gap length, is stop height, is core extension coil length, and is nonmagnetic bushing magnetoresistance.
The magnetic induction intensity B in the most saturated section of the core can be expressed by the size of the magnetic conductor and coil:
The volume of the electromagnet can be expressed by geometric size:
To make the optimization results is universal, the basic geometric dimensions of solenoid electromagnet are expressed by dimensionless relative values. The core diameter is taken as the basic size.
Then, the relative values of the electromagnetic force and the maximum magnetic induction intensity in the core can be written in the following forms:
The volume of the electromagnet is regarded as the objective function and expressed by relative size:
Among them, , can be expressed by , :
The sequential quadratic programming (SQP) algorithm is an effective algorithm for solving constrained optimization problems in small- and medium-sized programming. The full name of SQP algorithm is sequential quantitative programming, which is used to solve quadratic linear optimization problems with equality and inequality constraints and find a local optimal solution that meets the constraints. It belongs to the method under the discipline classification of numerical optimization. At a certain approximate solution of the problem, several quadratic programming subproblems are solved, and then a better approximate solution through the solutions of these subproblems is found. The method is used to optimize the structural parameters. In MATLAB, SQP is realized in three steps: updating the Hessian matrix of the Lagrange function; solving quadratic programming problems; and one-dimensional search and calculation of the objective function.
Independent variables: .
Objective function: .
Constraints: ; .
In the optimization calculation, the relevant parameters are set as shown in Table 1. The optimized values and the initial contrast values are shown in Figure 2. After optimizing the data, the volume of the target value-electromagnet part is 2857 mm3, which is 18.4% less than the theoretical value calculated from the initial value. This means that the miniaturization goal is effectively achieved.
The digital valve is an electric magnetic mechanical liquid multifield coupling miniature system, as shown in Figure 3. The voltage generates current through the circuit, and the current generates electromagnetic energy in the magnetic circuit, which overcomes the resistance of friction, hydrodynamic force, and so on, promotes the valve core to move, and changes the flow field.
The miniature high-speed digital valve is shown in Figure 4. The detailed structure can be found in reference [4].
The optimized parameters of the miniature high-speed digital valve are shown in Table 2.
3. Thermodynamic Analysis Theory
In the Workbench simulation platform, thermodynamic analysis mainly includes steady-state thermal equilibrium and dynamic heat transfer. The difference between thermal equilibrium and dynamic heat transfer is whether the temperature of the thermal field changes with time. The heat balance equation iswhere includes the conduction matrix of convection coefficient, thermal conductivity, radiation rate, and shape coefficient. is temperature vector, and is the heat flux vector. The simulation platform uses geometric parameters of the model, thermal performance parameters of materials, and boundary conditions to generate , and .
The temperature of the coil has an effect on the magnetic field strength and resistance. By establishing the thermodynamic field model of the valve, the temperature change of the coil when it works is obtained.
For the three-dimensional model, the temperature field satisfies the differential equation:where is the density of the material, is the specific heat capacity of the material, and , , and are the heat conductivity of the material along three directions. is the heat flux per unit volume in a unit time. In solving differential equations, initial conditions and boundary conditions need to be given. The initial conditions are the ambient temperature.
There are three kinds of boundary conditions. The first kind of boundary condition is that the temperature function on the boundary of an object is known and expressed aswhere is the boundary of the object, and is a known temperature function. The second kind of boundary condition is that the heat flux on the boundary of an object is known and expressed aswhere is the thermal conductivity, and is the heat flux function.
The third kind of boundary condition is the temperature and heat exchange coefficient of the fluid medium in contact with the object:
There are two parts of the heat source in the digital valve model. One is the Joule heat generated by the current through the electromagnetic coil, referred to as copper loss; the other is the eddy current and hysteresis loss in the magnet caused by the change of the current, and then heat, referred to as iron loss. The digital valve continuously inputs the voltage signal and produces Joule heat in the coil. The heat is transferred to the moving core and the static core through the coil skeleton, which changes the temperature of the magnetic field environment.
Copper loss heat generation model:
Heat generation model of iron loss:
In the formula, is the eddy current density, and is the conductivity.
Coil resistance varies with temperature. The relationship between temperature and resistance can be expressed as follows:
In the formula, is the resistance value of the coil at room temperature; is the temperature coefficient of the resistance.
4. Steady-State Simulation of Temperature Field
4.1. Finite Element Model
To improve computational efficiency, a half-model is built on the simulation platform. The digital valve gridding is shown in Figure 5. To calculate accurately, parts that have a great influence on heat transfer are encrypted separately. After meshing, the number of digital valve nodes is 112 7292, and the number of units is 364907.
Steady-state temperature field simulation needs to define material heat conductivity. The actual heat conductivity varies with the change in temperature. However, the temperature of the digital valve is relatively low in steady-state operation, and the range of change is small, which has limited impact on material heat conductivity. As a result, the constant heat conductivity of each material can be set in the simulation platform. The material parameters of each part are shown in Table 3.
Joule heat produced by electromagnetic coils acts as a heat source. In the simulation, Joule heat is applied to the simulation object as a thermal load. Because the thermal loads and boundary conditions vary with time and temperature, the thermal calculation is nonlinear. The distribution of the temperature field is shown in Figure 6. The lowest temperature is 33.368°C, and the highest temperature is 60.812°C. As shown in Figure 7, the maximum heat flux is 3.3352 W/mm2, and the minimum heat flux is 2.0569e-15 W/mm2. The heat flow distribution in the X direction is shown in Figure 8. The whole valve is affected by Joule heat, and the temperature distribution is nonuniform. The magnetic materials, such as magnetic conduction rings and static iron cores, are affected by heat, and the magnetic conductivity changes, so the solenoids of the whole valve will also change. Therefore, this kind of temperature conduction is not conducive to the performance of the valve. Firstly, temperature conduction in the high-speed digital valve will cause the electromagnetic characteristics of the electromagnetic material of the valve itself, which affects magnitude of the electromagnetic force, permeability, and magnetic field strength and thus affects the speed and displacement of the valve spool. Secondly, temperature conduction will also affect the viscosity characteristics of hydraulic oil, which results in leakage of high-speed digital valve and damage of the sealants. Thirdly, temperature conduction will lead to the imprecision of flow control. This study proposes a method of adding heat-insulating aluminum alloy around the coil to fix the heat source on the heat-insulating material-heat-insulating aluminum alloy. Avoid thermal diffusion and magnetic field changes all around.
The distribution of temperature field after heat insulation is shown in Figure 9, the total heat flux distribution is shown in Figure 10, and the X-direction heat flux distribution is shown in Figure 11. It can be seen that the temperature distribution after heat insulation is mostly concentrated near the heat source, the temperature change in other parts of the valve is small, and the magnetic field will hardly change.
5. Research on Electronic Control Board
The dynamic characteristics of the miniature high-speed digital valves are the result of the interaction and coupling of electromagnetism, machine, liquid, and heat. Among these, electricity is the source factor. The function of the driving circuit of the electronic control board is to convert the control signal PWM voltage into the current signal received by the digital valve, which directly determines the current characteristics of the digital valve and then determines the change trend of the electromagnetic force.
When compared with the traditional reverse unloading drive circuit, an improved reverse unloading drive electronic board using IR2301 chip to realize simultaneous switching control of two thyristors is designed in this study, as shown in Figure 12. The diagram shows that the electromagnet charges when the two thyristors are turned on and achieve stability quickly. Unloading and current returning to zero can accelerate the current fall speed of solenoid coil of the digital valves and improve the dynamic response characteristics.
The reverse unloading circuit mainly consists of the IR2301 chip, MOS transistor, ultra-high speed Schottky diode and current sampling chip AD8418. The principle of reverse unloading: when two MOS transistors are turned on at the same time, the power supply charges the solenoid coil forward rapidly. The current acquisition module adopts a zero drift, high resolution current detection amplifier AD8418. It has high precision under the input common-mode voltage of PWM output class, and the typical value of the input offset voltage is ±200 μV. The output range is 0∼5 V. When the output voltage is 0, the current is 0. Because the back-end circuit needs to collect the current value, it needs a low noise power supply as the supply voltage of current amplifier AD8418. A LT1761 positive linear regulator is selected in the circuit to ensure that the noise of the output current acquisition value is low. The electronic control board is shown in Figure 13.
The test platform for the electronic board is shown in Figure 14. The control signal is changed by adjusting the frequency and duty cycle of the PWM signal. In the test, the control signal with a frequency of 100 Hz and a duty cycle of 70% is selected. The sampling rate is 5000, the number of samples is 500, and the sampling time is 0.1 s. The current response curve is shown in Figure 15. Because the circuit adopts a reverse unloading method, the current can rapidly rise and fall along a straight line. The current rises and falls instantaneously along with change of working voltage, and there is almost no delay in time. As a result, the electronic board can be regarded as the output of a proportional link.
6. Conclusion
Digital hydraulics is a significant area of hydraulic technology, and its performance is becoming increasingly comparable to that of proportional and servo technology. The performance of the digital hydraulic system depends on the performance of the single valve. The simulation research and experiments are performed to obtain certain data. The following are the major conclusions:(1)To be more competitive in terms of system volume, the volume of a single valve should have miniaturization characteristics. For this reason, the structural parameters are optimized by the SQP method. After optimizing the data, the target volume of the electromagnet part is 2857 mm3, which is 18.4% less than the theoretical calculation value, thereby achieving the miniaturization goal.(2)The simulation results show that, in the original structure, heat transfer affects the temperature of electromagnetic components and the magnetic conductivity of the entire electromagnetic field. This study proposes adding a type of heat insulation material to the original structure. The heat source is contained within the heat insulation material, and it is incapable of transferring heat to the surrounding parts. The magnetic conductivity of the entire electromagnetic field is not affected. The simulation results show that the proposed method is both accurate and feasible, as well as producing reasonable outputs.(3)To accommodate the high-speed characteristics of the miniature digital valve, an improved reverse unloading drive electronic board has been developed, which uses an IR2301 chip to simultaneously control two thyristors. By setting up an electronic board test bench, the delay time between current rising and falling speeds in PWM wave control can be ignored. It satisfies the criteria for the high-speed characteristic of an electronic control board.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
The author declares that there are no conflicts of interest.
Acknowledgments
This work was supported by National Natural Science Foundation of China (52105041), Opening Fund of State Key Laboratory of Mechanical Transmission (SKLMT-MSKFKT-202018), Key Projects of Natural Science Research in Anhui Universities (KJ2020A0258), and Opening Fund of Engineering Research Center of Hydraulic Vibration and Control, Ministry of Education (HVC202002).