A control method for cooling water circulation system of multi-single crystal furnace

Abstract

The present invention provides a control method for a multi-single crystal furnace cooling water circulation system, comprising: establishing a multi-objective function of crystal growth rate and cooling water system energy consumption; solving the multi-objective function based on a non-dominated sorting genetic algorithm to obtain a target furnace temperature and a target cooling water flow rate; using fuzzy control theory to establish a fuzzy controller for temperature and flow rate regulation, and outputting a temperature regulation coefficient and a flow rate regulation coefficient through the fuzzy controller; a fan controls the temperature of a cooler according to the temperature regulation coefficient, and a motor controls the flow rate of a pump according to the flow rate regulation coefficient. The present invention can reduce the energy consumption of a cooling water circulation system and improve control accuracy.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a method for controlling a cooling water circulation system of a multi-single crystal furnace. Background Art

[0002] Single crystal silicon is a basic material in the electronic information industry. According to different crystal growth methods, it is divided into Czochralski method, zone melting method and epitaxial method. Among them, the Czochralski method has high purity, low defect density, low cost, and the growth diameter can reach more than several inches, which is the current development direction.

[0003] The CZ single crystal furnace needs to be cooled during the preparation process, which is generally achieved by a cooling water circulation system. Existing cooling water circulation systems mostly use a single-path water inlet to reach the cooling area, and after cooling is completed, it circulates from the water outlet to the water inlet. During the operation of the single crystal furnace, due to the high temperature working environment and long-term continuous operation, the cooling water system needs to bear a huge cooling load. Traditional cooling water systems may have problems such as high energy consumption and unsatisfactory cooling effect, which not only increases production costs, but may also have an adverse effect on the performance and life of the single crystal furnace.

[0004] At present, in order to reduce energy consumption, it is often achieved by optimizing the layout of cooling water pipelines and adopting high-efficiency cooling equipment. However, due to limitations in furnace safety and equipment performance, the overall energy consumption of the cooling water circulation system cannot be effectively improved, and the control accuracy of the current cooling water system is also relatively low. Summary of the invention

[0005] The object of the present invention is to provide a method for controlling a cooling water circulation system of a multi-single crystal furnace, so as to reduce the energy consumption of the cooling water circulation system and improve the control accuracy.

[0006] A control method for a cooling water circulation system of a multi-single crystal furnace, the cooling water circulation system of the multi-single crystal furnace comprising a plurality of single crystal furnaces, the water inlet of any single crystal furnace being connected to a cooler via a pump, the cooler being connected to a process water pool via a throttle valve, the cooler controlling the temperature via a fan, the pump controlling the flow rate via a motor, the water outlet of any single crystal furnace being connected to the process water pool via a switch valve, a sensor module being arranged on a circulation pipeline between any single crystal furnace and the process water pool, the sensor module being used to detect parameter information, the sensor module comprising a temperature sensor and a flow rate sensor, the temperature sensor being used to detect a cooling water inlet temperature and a cooling water outlet temperature, the flow rate sensor being used to detect a cooling water flow rate;

[0007] The control method comprises the following steps:

[0008] Step S1, establishing a multi-objective function of crystal growth rate and cooling water system energy consumption based on parameter information obtained by the sensor module;

[0009] Step S2, solving the multi-objective function based on a non-dominated sorting genetic algorithm to obtain a target furnace temperature and a target cooling water flow rate;

[0010] Step S3, according to the target furnace temperature, calculating the first temperature error value between the cooling water inlet temperature of the single crystal furnace and the target furnace temperature, and the second temperature error value between the cooling water outlet temperature and the target furnace temperature, according to the target cooling water flow rate, calculating the flow rate error value between the actual cooling water flow rate of the single crystal furnace and the target cooling water flow rate, based on the first temperature error value, the second temperature error value and the flow rate error value, using fuzzy control theory to establish a fuzzy controller for temperature and flow rate regulation, and outputting a temperature regulation coefficient and a flow rate regulation coefficient through the fuzzy controller;

[0011] Step S4, the fan controls the temperature of the cooler according to the temperature adjustment coefficient, and the motor controls the flow rate of the pump according to the flow rate adjustment coefficient.

[0012] According to the control method for the cooling water circulation system of multiple single crystal furnaces provided by the present invention, a multi-objective function of crystal growth rate and cooling water system energy consumption is first established, and then the multi-objective function is solved based on a non-dominated sorting genetic algorithm to obtain a target furnace temperature and a target cooling water flow rate, and then a fuzzy controller for temperature and flow rate regulation is established using fuzzy control theory. The fuzzy controller outputs a temperature regulation coefficient and a flow rate regulation coefficient, which has strong robustness. The temperature of the cooler is controlled by a fan according to the temperature regulation coefficient, and the flow rate of the pump is controlled by a motor according to the flow rate regulation coefficient, thereby improving the cooling control accuracy of the single crystal furnace and ensuring the stability of the cooling temperature. The method can simultaneously regulate the temperature and flow rate of cooling water for multiple single crystal furnaces, which helps to reduce energy consumption and maximize the production efficiency of the single crystal furnace. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a structural schematic diagram of the cooling water circulation system of the multi-single crystal furnace of the present invention;

[0014] Figure 2 The present invention is a flow chart of a method for controlling a multi-single crystal furnace cooling water circulation system. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] See also Figure 1The cooling water circulation system of the multi-single crystal furnace includes multiple single crystal furnaces, namely the first single crystal furnace, the second single crystal furnace, ..., the G Single crystal furnace, G Indicates the total number of single crystal furnaces. The water inlet of any single crystal furnace is connected to the cooler through a pump, the cooler is connected to the process water pool through a throttle valve, the cooler controls the temperature through a fan, the pump controls the flow rate through a motor, and the water outlet of any single crystal furnace is connected to the process water pool through a switch valve. The process water pool is equipped with an air cooling device. There is an impurity precipitation tank in the process water pool. By adding a coagulant, the impurities are accelerated to precipitate, and filters are installed at the outlet and inlet of the process water pool to prevent impurities from entering the cooling water circuit and causing blockage.

[0017] Specifically, the water inlet of the first single crystal furnace is connected to cooler one through pump one, cooler one is connected to the process water pool through throttle valve one, cooler one controls the temperature through the first fan, pump one controls the flow rate through the first motor, and the water outlet of the first single crystal furnace is connected to the process water pool through switch valve one. The water inlet of the second single crystal furnace is connected to cooler two through pump two, cooler two is connected to the process water pool through throttle valve two, cooler two controls the temperature through the second fan, pump two controls the flow rate through the second motor, and the water outlet of the second single crystal furnace is connected to the process water pool through switch valve two. G The water inlet of the single crystal furnace is pumped G With cooler G Connection, cooler G Through the throttle valve G Connected to process water tank, cooler G Through the G A fan controls the temperature, a pump G Through the G The motor controls the flow rate, G The water outlet of the single crystal furnace is connected through a switch valve G Connected to process water tank.

[0018] A sensor module is provided on the circulation pipeline of any single crystal furnace and process water pool. The sensor module is used to detect parameter information. The sensor module includes a temperature sensor and a flow rate sensor. The temperature sensor is used to detect the cooling water inlet temperature and the cooling water outlet temperature. The flow rate sensor is used to detect the cooling water flow rate. Figure 1 middle, T in1 is the cooling water inlet temperature of the first single crystal furnace, T out1 is the cooling water outlet temperature of the first single crystal furnace, T in2 is the cooling water inlet temperature of the second single crystal furnace, T out2 is the cooling water outlet temperature of the second single crystal furnace,T inG For the G The cooling water inlet temperature of a single crystal furnace, T outG For the G The cooling water outlet temperature of a single crystal furnace.

[0019] Based on the above multi-single crystal furnace cooling water circulation system, please refer to Figure 2 The present invention provides a method for controlling a cooling water circulation system of a multi-single crystal furnace, comprising steps S1 to S4:

[0020] Step S1, based on the parameter information obtained by the sensor module, a multi-objective function of crystal growth rate and cooling water system energy consumption is established.

[0021] Among them, taking the first single crystal furnace among multiple single crystal furnaces as an example, for the first single crystal furnace among multiple single crystal furnaces, a mathematical model of crystal growth rate and furnace temperature is established, and the expression is as follows:

[0022] ;

[0023] in, V Indicates the crystal growth rate; A is the crystal structure coefficient, specifically the coefficient related to the crystal structure, solute properties, etc.; represents the activation energy associated with the crystal growth mechanism, R is the gas constant, T is the furnace temperature, e is the base of natural logarithms.

[0024] The furnace temperature is related to the cooling water inlet temperature, outlet temperature and cooling water flow rate, which directly reflects the heat and efficiency taken away by the cooling water circulation system. The mathematical expression between the furnace temperature and the cooling water inlet temperature, outlet temperature and cooling water flow rate can be fitted through the state parameters of the actual single crystal furnace system, specifically:

[0025]

[0026] f represents the furnace temperature function, T in1 is the cooling water inlet temperature of the first single crystal furnace, T out1 is the cooling water outlet temperature of the first single crystal furnace, V w1 is the actual cooling water flow rate of the first single crystal furnace.

[0027] Determine the energy consumption of the cooling water section in the first single crystal furnace Q w , the expression is:

[0028] ;

[0029] in, is the data collection time interval, is the specific heat capacity corresponding to the cooling water inlet temperature of the first single crystal furnace, is the ambient temperature, is the specific heat capacity of cooling water at ambient temperature;

[0030] Establishing a multi-objective function of crystal growth rate and cooling water system energy consumption J , the expression is:

[0031] ;

[0032] Among them, max represents the maximum value function, and min represents the minimum value function.

[0033] Step S2, solving the multi-objective function based on the non-dominated sorting genetic algorithm to obtain the target furnace temperature and the target cooling water flow rate.

[0034] Among them, when solving the multi-objective function, the influence of subjective factors and objective factors are considered.

[0035] Step S2 specifically includes:

[0036] The expert scoring method is used to obtain the subjective weight of the optimization target in the multi-objective function R n , the expression is:

[0037] ;

[0038] in, For the Experts on n The score of the optimization goal, For the The authority coefficient of the experts;

[0039] Determine the evaluation indicator matrix Z , the expression is:

[0040] ;

[0041] in, is the first m Row, No. n The elements of the column, t represents the number of Pareto optimal solutions obtained, ;

[0042] Evaluation indicator matrixZ Each element in is standardized, and the expression is as follows:

[0043] ;

[0044] in, Represents the first m Row, No. n The elements of the column, is the maximum value of each column element;

[0045] Using C-OWA operator to calculate the objective weight of optimization objectives in multi-objective functions E n , the expression is:

[0046] ;

[0047] ;

[0048] in, is the objective weight after standardization, Indicates from t -1 element is randomly selected m -1 number of combinations of elements;

[0049] According to subjective weight R n and objective weight E n Get the comprehensive weight K n , the expression is:

[0050] K n = h 1 R n + h 2 E n ;

[0051] in, h 1 and h 2 is the linear combination coefficient;

[0052] Evaluation indicator matrix Z The weighted elements of are weighted to obtain the weighted matrix P , the expression is as follows:

[0053] ;

[0054] ;

[0055] in, is the weight matrix m Row, No. n Elements of a column;

[0056] The weight matrix P The smallest element in each column is the optimal solution , the largest element is the worst solution , to form a Pareto optimal solution set, the expression is as follows:

[0057] ;

[0058] in, p 1n is the first row and the n The elements of the column, p 2n is the second row and the n Elements of a column.

[0059] The Pareto optimal solution set includes at least a target furnace temperature and a target cooling water flow rate.

[0060] Furthermore, the Pareto optimal solution set is traversed, and the target furnace temperature and target cooling water flow rate are obtained by calculating the distance between each element in the weighted matrix and the optimal solution and the worst solution.

[0061] Specifically, the following formula is used to calculate the distance between each element in the weight matrix and the optimal solution and the worst solution:

[0062]

[0063] in, express p mn With the optimal solution The distance between express p mn and the worst solution The distance between

[0064] Calculate the proximity index of each optimal solution in the Pareto optimal solution set R m , the expression is as follows:

[0065] ;

[0066] The closeness index to each optimal solution R m Sort in descending order, R m The larger it is, the closer it is to the optimal distance.

[0067] Therefore, it will be close to the index R m The maximum optimal solution is taken as the final solution result of the multi-objective function, and the target furnace temperature of the first single crystal furnace is obtained from the final solution result. T t1 and target cooling water flow rate V tw1 .

[0068] Step S3, according to the target furnace temperature, calculate the first temperature error value between the cooling water inlet temperature of the single crystal furnace and the target furnace temperature, as well as the second temperature error value between the cooling water outlet temperature and the target furnace temperature; according to the target cooling water flow rate, calculate the flow rate error value between the actual cooling water flow rate of the single crystal furnace and the target cooling water flow rate; based on the first temperature error value, the second temperature error value and the flow rate error value, use fuzzy control theory to establish a fuzzy controller for temperature and flow rate regulation, and output the temperature regulation coefficient and the flow rate regulation coefficient through the fuzzy controller.

[0069] Wherein, step S3 specifically includes:

[0070] For the first single crystal furnace, calculate the error value between the cooling water inlet temperature of the first single crystal furnace and the first temperature in the target furnace , and the second temperature error value between the cooling water outlet temperature and the target furnace temperature , and according to the first temperature error value and the second temperature error value Get the temperature error value of the first single crystal furnace , the expression is as follows:

[0071] ;

[0072] ;

[0073] ;

[0074] Calculate the error between the actual cooling water flow rate and the target cooling water flow rate of the first single crystal furnace , the expression is:

[0075] ;

[0076] The designed fuzzy rule table will and As input, the temperature adjustment factor K T and flow rate adjustment factor K VAs the output, the Mamdani fuzzy reasoning method is used, and five language variables are used to describe it. The five language variables are: negative big (NB), negative small (NS), zero (ZE), positive small (PS), and positive big (PB). The membership function selects trigonometric function, and the centroid method is used to defuzzify. Finally, the fuzzy reasoning result is converted into the temperature adjustment coefficient. K T and flow rate adjustment factor K V .

[0077] Specifically, the fuzzy rule table of temperature adjustment coefficient is shown in Table 1:

[0078] Table 1

[0079]

[0080] The fuzzy rule table of flow rate adjustment coefficient is shown in Table 2:

[0081] Table 2

[0082]

[0083] In this embodiment, the fuzzy reasoning process satisfies the following conditional formula:

[0084] ;

[0085] in, is the output control quantity of the fuzzy controller, which includes the temperature adjustment coefficient K T and flow rate adjustment factor K V ; is the center of the fuzzy set generated according to the fuzzy rule table, Fuzzy rules k The area under the corresponding membership function is, h is the number of output variables.

[0086] Step S4, the fan controls the temperature of the cooler according to the temperature adjustment coefficient, and the motor controls the flow rate of the pump according to the flow rate adjustment coefficient.

[0087] In specific implementation, when the temperature error value Greater than the temperature threshold When the temperature error is Less than the temperature threshold When the temperature of the cooler is increased, the fan is adjusted according to the temperature adjustment coefficient, and the motor is adjusted according to the flow rate adjustment coefficient to reduce the speed of the pump, thereby ensuring the cooling control accuracy of the single crystal furnace and the stability of the cooling temperature.

[0088] In addition, in this embodiment, the sensor module further includes a pressure sensor, and the pressure sensor is used to detect the cooling water inlet pressure and the cooling water outlet pressure;

[0089] The control method further comprises:

[0090] When the difference between the cooling water inlet pressure and the cooling water outlet pressure is greater than the pressure threshold, e.g. V out - V in > V 0 o'clock, V out Indicates the cooling water outlet pressure, V in Indicates the cooling water inlet pressure, V 0 indicates the pressure threshold, which indicates that the cooling water circulation system of the multi-crystal furnace has a fault. At this time, the switch valve is closed. It is necessary to notify the staff to check, lower the furnace temperature and reduce the cooling water flow.

[0091] In summary, according to the control method of the cooling water circulation system of multiple single crystal furnaces provided by the present invention, a multi-objective function of crystal growth rate and cooling water system energy consumption is first established, and then the multi-objective function is solved based on the non-dominated sorting genetic algorithm to obtain the target furnace temperature and the target cooling water flow rate, and then the fuzzy control theory is used to establish a fuzzy controller for temperature and flow rate regulation. The fuzzy controller outputs the temperature regulation coefficient and the flow rate regulation coefficient, which has strong robustness. The fan controls the temperature of the cooler according to the temperature regulation coefficient, and the motor controls the flow rate of the pump according to the flow rate regulation coefficient, which improves the cooling control accuracy of the single crystal furnace and ensures the stability of the cooling temperature. This method can simultaneously regulate the temperature and flow rate of cooling water for multiple single crystal furnaces, which helps to reduce energy consumption and maximize the production efficiency of the single crystal furnace.

[0092] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0093] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for controlling a cooling water circulation system of a multi-single crystal furnace, characterized in that: The multi-single crystal furnace cooling water circulation system includes multiple single crystal furnaces, the water inlet of any single crystal furnace is connected to the cooler through a pump, the cooler is connected to the process water pool through a throttle valve, the cooler controls the temperature through a fan, the pump controls the flow rate through a motor, the water outlet of any single crystal furnace is connected to the process water pool through a switch valve, and a sensor module is provided on the circulation pipeline of any single crystal furnace and the process water pool, the sensor module is used to detect parameter information, the sensor module includes a temperature sensor and a flow rate sensor, the temperature sensor is used to detect the cooling water inlet temperature and the cooling water outlet temperature, and the flow rate sensor is used to detect the cooling water flow rate; The control method comprises the following steps: Step S1, establishing a multi-objective function of crystal growth rate and cooling water system energy consumption based on parameter information obtained by the sensor module; Step S2, solving the multi-objective function based on a non-dominated sorting genetic algorithm to obtain a target furnace temperature and a target cooling water flow rate; Step S3, according to the target furnace temperature, calculating the first temperature error value between the cooling water inlet temperature of the single crystal furnace and the target furnace temperature, and the second temperature error value between the cooling water outlet temperature and the target furnace temperature, according to the target cooling water flow rate, calculating the flow rate error value between the actual cooling water flow rate of the single crystal furnace and the target cooling water flow rate, based on the first temperature error value, the second temperature error value and the flow rate error value, using fuzzy control theory to establish a fuzzy controller for temperature and flow rate regulation, and outputting a temperature regulation coefficient and a flow rate regulation coefficient through the fuzzy controller; Step S4, the fan controls the temperature of the cooler according to the temperature adjustment coefficient, and the motor controls the flow rate of the pump according to the flow rate adjustment coefficient; Step S1 specifically includes: For the first single crystal furnace among multiple single crystal furnaces, a mathematical model of crystal growth rate and furnace temperature is established, and the expression is as follows: ； ； in, V represents the crystal growth rate, A is the crystal structure coefficient, represents the activation energy associated with the crystal growth mechanism, R is the gas constant, T is the furnace temperature, e is the base of natural logarithms, f represents the furnace temperature function, T in1 is the cooling water inlet temperature of the first single crystal furnace, T out1 is the cooling water outlet temperature of the first single crystal furnace, V w1 is the actual cooling water flow rate of the first single crystal furnace; Determine the energy consumption of the cooling water section in the first single crystal furnace Q w , the expression is: ； in, is the data collection time interval, is the specific heat capacity corresponding to the cooling water inlet temperature of the first single crystal furnace, is the ambient temperature, is the specific heat capacity of cooling water at ambient temperature; Establishing a multi-objective function of crystal growth rate and cooling water system energy consumption J , the expression is: ； Among them, max represents the maximum value function, and min represents the minimum value function; Step S2 specifically includes: The expert scoring method is used to obtain the subjective weight of the optimization target in the multi-objective function R n , the expression is: ； in, For the Experts on n The score of the optimization goal, For the The authority coefficient of the experts; Determine the evaluation indicator matrix Z , the expression is: ； in, is the first m Row, No. n The elements of the column, t represents the number of Pareto optimal solutions obtained, ; Evaluation indicator matrix Z Each element in is standardized, and the expression is as follows: ； in, Represents the first m Row, No. n The elements of the column, is the maximum value of each column element; Using C-OWA operator to calculate the objective weight of optimization objectives in multi-objective functions E n , the expression is: ； ； in, is the objective weight after standardization, Indicates from t -1 element is randomly selected m -1 number of combinations of elements; According to subjective weight R n and objective weight E n Get the comprehensive weight K n , the expression is: K n = h 1 R n + h 2 E n ； in, h 1 and h 2 is the linear combination coefficient; Evaluation indicator matrix Z The weighted elements of are weighted to obtain the weighted matrix P , the expression is as follows: ； ； in, is the weight matrix m Row, No. n Elements of a column; The weight matrix P The smallest element in each column is the optimal solution , the largest element is the worst solution , to form a Pareto optimal solution set, the expression is as follows: ； in, p 1n is the first row and the n The elements of the column, p 2n is the second row and the n Elements of a column; Traverse the Pareto optimal solution set, and obtain the target furnace temperature and target cooling water flow rate by calculating the distance between each element in the weighted matrix and the optimal solution and the worst solution; By calculating the distance between each element in the weighted matrix and the optimal solution and the worst solution, the target furnace temperature and the target cooling water flow rate are obtained, including: The following formula is used to calculate the distance between each element in the weight matrix and the optimal solution and the worst solution: in, express p mn With the optimal solution The distance between express p mn and the worst solution The distance between Calculate the proximity index of each optimal solution in the Pareto optimal solution set R m , the expression is as follows: ； Will approach the index R m The maximum optimal solution is taken as the final solution result of the multi-objective function, and the target furnace temperature of the first single crystal furnace is obtained from the final solution result. T t1 and target cooling water flow rate V tw1 ; Step S3 specifically includes: For the first single crystal furnace, calculate the error value between the cooling water inlet temperature of the first single crystal furnace and the first temperature in the target furnace , and the second temperature error value between the cooling water outlet temperature and the target furnace temperature , and according to the first temperature error value and the second temperature error value Get the temperature error value of the first single crystal furnace , the expression is as follows: ； ； ； Calculate the error between the actual cooling water flow rate and the target cooling water flow rate of the first single crystal furnace , the expression is: ； The designed fuzzy rule table will and As input, the temperature adjustment factor K T and flow rate adjustment factor K V As the output, the Mamdani fuzzy reasoning method is used, and five language variables are used to describe it. The five language variables are: negative large, negative small, zero, positive small, and positive large. The membership function selects trigonometric function, and the centroid method is used to defuzzify. Finally, the fuzzy reasoning result is converted into the temperature adjustment coefficient. K T and flow rate adjustment factor K V ; Step S3 satisfies the following conditional formula: ； in, is the output control quantity of the fuzzy controller, which includes the temperature adjustment coefficient K T and flow rate adjustment factor K V ; is the center of the fuzzy set generated according to the fuzzy rule table, Fuzzy rules k The area under the corresponding membership function is, h is the number of output variables.

2. The method for controlling the cooling water circulation system of a multi-single crystal furnace according to claim 1, characterized in that: The sensor module also includes a pressure sensor, which is used to detect the cooling water inlet pressure and the cooling water outlet pressure; The control method further comprises: When the difference between the cooling water inlet pressure and the cooling water outlet pressure is greater than the pressure threshold, it is determined that the multi-single crystal furnace cooling water circulation system has a fault, and the switch valve is closed at this time.

Images