Electric heating device parameter optimization method based on multi-objective analytic hierarchy process and genetic algorithm

Abstract

The invention relates to the field of electric heating device parameter optimization, in particular to an electric heating device parameter optimization method based on multi-objective analytic hierarchy process and a genetic algorithm. According to the scheme, a hierarchical evaluation model is constructed, the hierarchical evaluation model comprises a target layer, a criterion layer and a measure layer, the target layer is final reflection of indexes of all the layers, and the criterion layer comprises factors influencing device performance; the factors comprise electric heating device output line voltage, electric heating device output harmonic content, electric heating device output heating power, electric heating device efficiency, electric heating device torque pulsation and electric heating device overall power density, and the measure layer comprises measures adopted for adjusting electric heating device parameters; and then constructing a judgment matrix, screening out parameters which have the greatest influence on the key performance indexes of the device from the structural parameters of the electric heating device, and then comprehensively optimizing the key performance indexes of the device by utilizing a genetic algorithm based on the screened parameters to obtain optimal parameters. The method is suitable for parameter optimization of the electric heating device.

Description

Technical Field

[0001] The invention relates to the field of electric heating device parameter optimization, and in particular to an electric heating device parameter optimization method based on multi-objective hierarchical analysis and genetic algorithm. Background Art

[0002] The wind thermal system converts all wind energy into heat energy for reuse, effectively solving the problem of intermittent and volatile wind energy, and has good economic efficiency. However, this technical solution has low overall efficiency, and heat energy cannot be transmitted over long distances, which cannot adapt to the long distance between energy production and consumption in my country.

[0003] Therefore, the current practice is to consider using an electric-thermal multi-energy flexible magnetic control power generation technology that can flexibly convert wind energy into both electrical energy and thermal energy. This technology achieves the simultaneous conversion of wind energy into electrical energy and thermal energy through an electric-thermal multi-energy flexible magnetic control power generation device. This technology first meets the energy demand in the power grid, and then converts excess wind energy into thermal energy for storage or utilization. This solution can effectively improve the overall efficiency of the system while solving the problem of wind energy utilization.

[0004] However, there is currently a lack of technical means to optimize the parameters of the electric heating device in the above scheme, resulting in low overall performance of the electric heating device. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for optimizing the parameters of an electric heating device based on multi-objective hierarchical analysis and genetic algorithm, thereby optimizing the parameters of the electric heating device and improving the comprehensive performance of the electric heating device.

[0006] The present invention adopts the following technical scheme to achieve the above-mentioned purpose. The present invention provides a method for optimizing parameters of an electric heating device based on multi-objective hierarchical analysis and genetic algorithm, comprising:

[0007] S1. Construct a hierarchical evaluation model;

[0008] The hierarchical evaluation model includes a target layer, a criterion layer and a measure layer. The target layer is the final reflection of the indicators of each layer, that is, the performance indicators of the electric heating device reach the comprehensive optimum. The criterion layer includes factors that affect the performance of the device, including the output line voltage of the electric heating device, the output harmonic content of the electric heating device, the output heating power of the electric heating device, the efficiency of the electric heating device, the torque ripple of the electric heating device and the overall power density of the electric heating device. The measure layer includes measures taken to adjust the parameters of the electric heating device.

[0009] S2, construct judgment matrix;

[0010] The judgment matrix determines the weight of each element to an element in the upper layer by comparing each element with each other. The values ​​in the judgment matrix are given by Santy's 1-9 scaling method. The 1-9 scale represents the importance of two elements compared with each other, where 1 element represents that the two elements have the same importance. The larger the scale, the higher the importance. The scaling method uses the reciprocal of 1-9 to represent the importance of two elements when the order is swapped.

[0011] For the elements of any layer, a judgment matrix is ​​established as follows:

[0012]

[0013] In the formula, a ij Represents the importance of the i-th element compared to the j-th element. The elements in A satisfy:

[0014] a ij >0

[0015]

[0016] a ii =1

[0017] S3. Solve the weight of each indicator through the judgment matrix;

[0018] Calculate the mth power of the product of each row of the judgment matrix to obtain an m-dimensional vector. The m-dimensional vector is the weight of each indicator before normalization. m is the row and column information of the judgment matrix. The number of rows and columns of the judgment matrix is ​​equal. The calculation formula is as follows:

[0019]

[0020] The obtained data is standardized to obtain the weight vector, and the calculation formula is as follows:

[0021]

[0022] S4, performing consistency check on the calculation result of the weight vector;

[0023] Calculate the maximum characteristic root and CI value, where the calculation formula of the maximum characteristic root satisfies:

[0024]

[0025] In the formula, λ max is the maximum eigenvalue of the judgment matrix, AW is the accumulated value of the judgment matrix multiplied by the standardized weights and then row by row;

[0026] The calculation formula of CI value is as follows:

[0027]

[0028] After obtaining the CI value, the results are checked for consistency in combination with the RI value table. The checking method is to calculate the CR value. When the CR value is less than the set threshold, it is judged to have passed the consistency check. The CR value calculation formula is as follows:

[0029] RI is a statistical parameter, which can be directly obtained by looking up the RI value table;

[0030] S5. Calculate the total hierarchical ranking weight. The calculation formula is as follows:

[0031]

[0032] In the formula, w is the total weight, w bai Represents the bottom layer b method to the middle layer a i The impact weight of the target, w aiz Represents the middle layer a i The impact weight of the target on the total target of z;

[0033] S6. Filter out the parameters of the electric heating device that have the greatest impact on the performance index of the electric heating device according to the total weight;

[0034] S7. Based on the screened parameters of the electric heating device, the performance indicators of the electric heating device are optimized according to the NSGA-II genetic algorithm.

[0035] Furthermore, step S7 specifically includes:

[0036] S71, initializing the population, that is, determining the decision variables and objective function values ​​of each individual, wherein the individual is the screened electric heating device parameter, and the objective function value is the set performance index value;

[0037] S72, sort the individuals in the population by non-dominance, divide the individuals into different levels according to the dominance relationship, and calculate the dominance number n p and the dominated set S p to judge;

[0038] S73, calculating the crowding degree of each individual to maintain the diversity of the population. The crowding degree refers to the distance between an individual and its adjacent individuals in the target space.

[0039] S74. Select the next generation population based on non-dominated sorting and crowding distribution;

[0040] S75, steps S72 to S71, until the termination condition is met.

[0041] Furthermore, step S72 specifically includes:

[0042] S701, if the dominance number is 0, then the solution p is on the first level non-dominated frontier, find the dominance number and dominated set of each solution and the first level non-dominated frontier;

[0043] S702, for each solution p with a dominance number of 0, traverse each member q in the dominance set of p, and reduce the dominance number of q by 1. If any q has a dominance number of 0, put it into a separate list Q, and the members in Q are the second-level non-dominated frontier;

[0044] S703. Repeat step S702 for each member in Q to find the third layer, and then repeat the process from steps S701 to S702 until all solutions are sorted.

[0045] Furthermore, step S5 further includes:

[0046] The consistency test is performed on the total hierarchical sorting. Combined with the RI value table, the results are checked for consistency. The test method is to calculate the CR value. When the CR value is less than the set threshold, it is judged that it has passed the consistency test. The CR value calculation formula is as follows:

[0047]

[0048] Furthermore, the parameters of the electric heating device include:

[0049] The inner diameter of the power generation stator, the effective length of the device, the permanent magnet pole arc coefficient, the thickness in the magnetization direction, the stator slot size parameters, the air gap length of the power generation stator part, the air gap length of the heating stator part, the thickness of the heating stator, the number of holes in the heating stator, and the radius of the holes in the heating stator;

[0050] The inner diameter of the power generation stator, the effective length of the device, the pole arc coefficient of the permanent magnet of the power generation part and the thickness in the magnetization direction, as well as the stator slot size parameters affect the performance indicators of the power generation part;

[0051] The thickness of the heating stator, the radius of the hole of the heating stator, the pole arc coefficient of the permanent magnet of the heating part and the thickness in the magnetizing direction affect the performance indicators of the heating part.

[0052] The beneficial effects of the present invention are:

[0053] The present invention first screens out the influence of the device structural parameters on the device key performance indicators based on the multi-objective hierarchical analysis method, and then uses the NSGA-II genetic algorithm based on these key structural parameters to comprehensively optimize the device key performance indicators to obtain the optimal structural parameters under limited conditions. The present invention can optimize multiple objectives and achieve the comprehensive optimum under limited target conditions. The optimization of the parameters of the electric heating device is achieved, and the comprehensive performance of the electric heating device is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of a method for optimizing parameters of an electric heating device based on multi-objective hierarchical analysis and genetic algorithm provided by an embodiment of the present invention;

[0055] Figure 2 is a torque comparison curve diagram of an electric heating device before and after optimization provided by an embodiment of the present invention;

[0056] Figure 3 It is a comparison curve diagram of the heating power of the electric heating device before and after optimization provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0058] The electric heating device of the present invention can be divided into components such as a power generation stator, a heating stator, a rotor, and a permanent magnet according to the structure. The power generation and heating functions of the device can be realized respectively by the power generation stator and the heating stator. In particular, in order to increase the heating power and facilitate the transmission of heat energy, the heating stator part is often punched on the side close to the permanent magnet. The permanent magnet can establish the working magnetic flux of the device, and the alternating magnetic field can be generated by driving the permanent magnet to rotate through the rotor.

[0059] The structural parameters of the electric heating device mainly include: the inner and outer diameters of the power generation stator D o1 、Effective length of the device L ef 、Permanent magnet pole arc coefficient α P , thickness in the magnetization direction h m , stator slot size parameters, air gap length of the power generation stator part, air gap length of the heating stator part, heating stator thickness d is , the number of holes in the heating stator, the radius of the holes in the heating stator i wait.

[0060] Among them, the inner and outer diameters of the power generation stator, the effective length of the device, the pole arc coefficient of the permanent magnet of the power generation part and the thickness in the magnetization direction, as well as the stator slot size parameters affect the performance indicators of the power generation part; the thickness of the heating stator, the radius of the hollow circular hole of the heating stator, the pole arc coefficient of the permanent magnet of the heating part and the thickness in the magnetization direction affect the performance indicators of the heating part.

[0061] In the design process of electric heating devices, the performance indicators of the power generation part should be met first. Therefore, in general, the key performance indicators of electric heating devices mainly include: device output line voltage, device output harmonic content, device output heating power, device efficiency, device torque pulsation and device overall power density.

[0062] Based on this, the present invention provides a method for optimizing parameters of an electric heating device based on multi-objective hierarchical analysis and genetic algorithm, such as Figure 1 As shown, specifically including:

[0063] Screening of key parameters of electric heating device based on multi-objective analytic hierarchy process:

[0064] (1) First, a hierarchical evaluation model needs to be constructed. The model mainly includes a target layer, a criterion layer, and a measure layer. Among them, the target layer is the final reflection of the indicators of each layer. For the method of the present invention, the target layer is the device performance that achieves the comprehensive optimization; the criterion layer mainly includes the main factors affecting the device performance. The present invention mainly includes the device output line voltage, the device output harmonic content, the device output heating power, the device efficiency, the device torque pulsation, and the device overall power density; the measure layer mainly includes the measures that can be taken to achieve the goal. The present invention mainly adjusts the structural parameters of the electric heating device. The hierarchical evaluation model serves as the basis for the final calculation of the weight of the impact of each topological structure parameter change on the device performance. The total weight is obtained based on the comprehensive calculation of the weights between different layers of the hierarchical evaluation model.

[0065] (2) Construct a judgment matrix. The judgment matrix is ​​to determine the weight of each element relative to an element in the upper layer by comparing each element with each other. The values ​​in the judgment matrix are often given by Santy's 1-9 scaling method. The 1-9 scale represents the importance of two elements compared to each other, where 1 represents that the two elements have the same importance. The larger the scale, the higher the importance. The scaling method uses the reciprocal of 1-9 to represent the importance of two elements when the order of comparison is swapped.

[0066] Therefore, for the elements of any layer, the present invention can establish a judgment matrix as follows:

[0067]

[0068] In the formula, a ij Represents the importance of the i-th element compared to the j-th element. The elements in A satisfy:

[0069] a ij >0

[0070]

[0071] a ii =1

[0072] (3) The weights of each indicator are solved through the judgment matrix. This process is also called hierarchical single sorting. The specific steps include:

[0073] 1) Calculate the mth power of the product of each row of the judgment matrix to obtain an m-dimensional vector. This m-dimensional vector is the weight of each indicator before normalization. m is the row and column information of the judgment matrix. The number of rows and columns of the judgment matrix is ​​usually equal. The calculation formula is as follows:

[0074]

[0075] 2) The obtained data is standardized to obtain the weight vector, and the weight is obtained at this time. The calculation formula is as follows:

[0076]

[0077] (4) Perform consistency check on the weight calculation results. The specific steps include:

[0078] 1) First, the maximum characteristic root and CI value need to be calculated, where the calculation formula of the maximum characteristic root satisfies:

[0079]

[0080] In the formula, λ max is the maximum eigenvalue of the judgment matrix, and AW is the accumulated value of the judgment matrix multiplied by the standardized weights and then row by row.

[0081] After solving the maximum eigenvalue, the CI value can be obtained. The calculation formula is as follows:

[0082]

[0083] 2) After obtaining the CI value, the present invention can perform a consistency test on the result in combination with the RI value table. The test method is to obtain the CR value. When CR<0.1, it can be considered that it has passed the consistency test. The calculation formula is as follows:

[0084] RI is a statistical parameter and can be directly obtained by looking up the RI value table.

[0085] (5) After the above process is completed, the hierarchical total sorting process can be carried out. The purpose of the hierarchical total sorting is to reflect the weight of the lowest level to the highest level target. For the present invention, the purpose of the hierarchical total sorting is to screen out a group of topological structure parameters that have the greatest impact on the key performance indicators of the device, so as to facilitate the subsequent use of the NSGA-II algorithm to set the change parameters during optimization. The final hierarchical total sorting weight calculation formula is:

[0086]

[0087] In the formula, w is the total weight, which is obtained by comprehensive calculation of the weights between different layers of the hierarchical evaluation model. The total weight of the present invention characterizes the influence of each topological structure parameter on the key performance indicators of the device. Through the comprehensive ranking and comparison of the total weight, a group of topological structure parameters with the greatest influence on the key performance parameters of the device can be screened. bai Represents the bottom layer b method to the middle layer a i The impact weight of the target, w aiz Represents the middle layer a i The weight of the target's impact on the total target z.

[0088] The total hierarchical sorting also needs to be checked for consistency. The judgment index is consistent with the hierarchical single sorting. The verification method is to calculate the CR value. When the CR value is less than the set threshold, it is judged that it has passed the consistency test. The CR value calculation formula is as follows:

[0089]

[0090] According to the key topological structure parameters selected by the multi-objective hierarchical analysis method, the performance indicators of the electric heating device are optimized by the NSGA-II genetic algorithm. The specific steps are as follows:

[0091] (1) Initializing the population, i.e., determining the decision variables and objective function values ​​of each individual. Here, the individual is the device topology parameter selected by the multi-objective analytic hierarchy process, the objective function value is the set key performance indicator of the device, and the range of the individual decision variables is set based on the constraints and expected range of the topological structure;

[0092] (2) Perform non-dominated sorting on the individuals in the population and divide the individuals into different levels according to their dominance relationship. This is mainly done by calculating the dominance number n p and the dominated set S p To judge, the steps include:

[0093] 1) If the dominance number is 0, then the solution p is on the first level of non-dominated frontier. Through the above operations, the dominance number and dominated set of each solution and the first level of non-dominated frontier are found;

[0094] 2) Then, for each solution p with a dominance number of 0, traverse each member q in the dominance set of p, and reduce the dominance number of q by 1. If any q has a dominance number of 0, put it into a separate list Q. The members in Q are the second-level non-dominated frontier;

[0095] 3) Repeat 2) for each member in Q to find the third layer, and then repeat 1) and 2) until all solutions are found and the sorting is completed.

[0096] (3) Calculate the crowding degree of each individual to maintain the diversity of the population. Crowding degree refers to the distance between an individual and its neighboring individuals in the target space.

[0097] (4) According to the non-dominated sorting and crowding distribution, the next generation of population is selected. The principle of selection is to retain as many individuals as possible in the Pareto optimal solution set and maintain the diversity of the population.

[0098] (5) Repeat steps 2 to 4 until the termination condition is met.

[0099] In order to verify the feasibility of this optimization scheme, an electric-thermal multi-energy flexible magnetic control power generation device with a dual-stator configuration was built in the finite element simulation. The output line voltage, harmonic content, heating power, loss ratio within a reasonable range, as well as the minimum torque pulsation and the highest power density were the optimization goals of this paper. It was optimized based on this scheme. The comparison of key structural parameters of the device before and after optimization is shown in Table 1. The device in the embodiment adopts a trapezoidal slot structure.

[0100] Table 1 Comparison of key structural parameters of the device before and after optimization

[0101] parameter Before optimization After optimization Generator stator inner diameter / mm 170 167.1 Motor length / mm 65 70.938 External permanent magnet pole arc coefficient 0.6 0.674 External permanent magnet thickness / mm 6 3.925 <![CDATA[Stator slot h b > 3.6 3.535 <![CDATA[Dimension parameter b t > 0.5 0.505 <![CDATA[Number / mm h t1 > 3.5 3.373 Heating stator thickness / mm 10 13.397 Hole radius / mm 3.6 2.444 Internal permanent magnet pole arc coefficient 0.9 0.951 Inner permanent magnet thickness / mm 9 8.122

[0102] The changes of device torque and heating power before and after optimization are as follows Figure 2 , Figure 3 After optimization, the effective value of the line voltage is 380.08V, the THD value of the harmonic content is 0.499%, the heat power is 3321.2W, the torque ripple is 0.768%, the motor efficiency is 97.41%, and the power density is 2.5513×10 6 W / m 2 The heating power increased by 20.50% compared with the initial optimization, the overall efficiency of the motor increased slightly by 0.13%, and the power density increased by 13.60%, which verified the effectiveness of the optimization method. The comparison of parameters before and after optimization is shown in Table 2.

[0103] Table 2 Comparison of parameters before and after optimization

[0104]

[0105]

[0106] In summary, the present invention optimizes the key performance indicators of the selected device through the above scheme, and based on the selected important topological structure parameters, effectively improves the performance of the device and reaches the comprehensive optimum under limited conditions.

[0107] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for optimizing parameters of electric heating devices based on multi-objective hierarchical analysis and genetic algorithm, characterized in that: include: S1. Construct a hierarchical evaluation model; The hierarchical evaluation model includes a target layer, a criterion layer and a measure layer. The target layer is the final reflection of the indicators of each layer, that is, the performance indicators of the electric heating device reach the comprehensive optimum. The criterion layer includes factors that affect the performance of the device, including the output line voltage of the electric heating device, the output harmonic content of the electric heating device, the output heating power of the electric heating device, the efficiency of the electric heating device, the torque ripple of the electric heating device and the overall power density of the electric heating device. The measure layer includes measures taken to adjust the parameters of the electric heating device. S2, construct judgment matrix; The judgment matrix determines the weight of each element to an element in the upper layer by comparing each element with each other. The values ​​in the judgment matrix are given by Santy's 1-9 scaling method. The 1-9 scale represents the importance of two elements compared with each other, where 1 element represents that the two elements have the same importance. The larger the scale, the higher the importance. The scaling method uses the reciprocal of 1-9 to represent the importance of two elements when the order is swapped. For the elements of any layer, a judgment matrix is ​​established as follows: In the formula, a ij Represents the importance of the i-th element compared to the j-th element. The elements in A satisfy: a ij >0 a ii ＝1 S3. Solve the weight of each indicator through the judgment matrix; Calculate the mth power of the product of each row of the judgment matrix to obtain an m-dimensional vector. The m-dimensional vector is the weight of each indicator before normalization. m is the row and column information of the judgment matrix. The number of rows and columns of the judgment matrix is ​​equal. The calculation formula is as follows: The obtained data is standardized to obtain the weight vector, and the calculation formula is as follows: S4, performing consistency check on the calculation result of the weight vector; Calculate the maximum characteristic root and CI value, where the calculation formula of the maximum characteristic root satisfies: In the formula, λ max is the maximum eigenvalue of the judgment matrix, AW is the accumulated value of the judgment matrix multiplied by the standardized weights and then row by row; The calculation formula of CI value is as follows: After obtaining the CI value, the results are checked for consistency in combination with the RI value table. The checking method is to calculate the CR value. When the CR value is less than the set threshold, it is judged to have passed the consistency check. The CR value calculation formula is as follows: RI is a statistical parameter, which can be directly obtained by consulting the RI value table; S5. Calculate the total hierarchical ranking weight. The calculation formula is as follows: In the formula, w is the total weight, w bai Represents the bottom layer b method to the middle layer a i The impact weight of the target, w aiz Represents the middle layer a i The impact weight of the target on the total target of z; S6. Filter out the parameters of the electric heating device that have the greatest impact on the performance index of the electric heating device according to the total weight; S7. Based on the screened parameters of the electric heating device, the performance indicators of the electric heating device are optimized according to the NSGA-II genetic algorithm.

2. The method for optimizing parameters of electric heating device based on multi-objective hierarchical analysis and genetic algorithm according to claim 1 is characterized in that: Step S7 specifically includes: S71, initializing the population, that is, determining the decision variables and objective function values ​​of each individual, wherein the individual is the screened electric heating device parameter, and the objective function value is the set performance index value; S72, sort the individuals in the population by non-dominance, divide the individuals into different levels according to the dominance relationship, and calculate the dominance number n p and the dominated set S p to judge; S73, calculating the crowding degree of each individual to maintain the diversity of the population. The crowding degree refers to the distance between an individual and its adjacent individuals in the target space. S74. Select the next generation population based on non-dominated sorting and crowding distribution; S75, steps S72 to S71, until the termination condition is met.

3. The method for optimizing parameters of electric heating device based on multi-objective hierarchical analysis and genetic algorithm according to claim 2 is characterized in that: Step S72 specifically includes: S701, if the dominance number is 0, then the solution p is on the first level non-dominated frontier, and the dominance number and the dominated set of each solution and the first level non-dominated frontier are found; S702, for each solution p with a dominance number of 0, traverse each member q in the dominance set of p, and reduce the dominance number of q by 1. If any q has a dominance number of 0, put it into a separate list Q, and the members in Q are the second-level non-dominated frontier; S703. Repeat step S702 for each member in Q to find the third layer, and then repeat the process from steps S701 to S702 until all solutions are sorted.

4. The method for optimizing parameters of electric heating device based on multi-objective hierarchical analysis and genetic algorithm according to claim 1 is characterized in that: Step S5 also includes: The consistency test is performed on the total hierarchical sorting. Combined with the RI value table, the results are checked for consistency. The test method is to calculate the CR value. When the CR value is less than the set threshold, it is judged that it has passed the consistency test. The CR value calculation formula is as follows:

5. The method for optimizing parameters of electric heating device based on multi-objective hierarchical analysis and genetic algorithm according to claim 1 is characterized in that: The parameters of the electric heating device include: The inner diameter of the power generation stator, the effective length of the device, the permanent magnet pole arc coefficient, the thickness in the magnetization direction, the stator slot size parameters, the air gap length of the power generation stator part, the air gap length of the heating stator part, the thickness of the heating stator, the number of holes in the heating stator, and the radius of the holes in the heating stator; The inner diameter of the power generation stator, the effective length of the device, the pole arc coefficient of the permanent magnet of the power generation part and the thickness in the magnetization direction, as well as the stator slot size parameters affect the performance indicators of the power generation part; The thickness of the heating stator, the radius of the hole of the heating stator, the pole arc coefficient of the permanent magnet of the heating part and the thickness in the magnetizing direction affect the performance indicators of the heating part.

Images