Vacuum concentration data analysis system based on cloud computing

Abstract

The invention relates to the technical field of vacuum concentration data analysis, and discloses a cloud computing-based vacuum concentration data analysis system, which comprises an equipment layer, a control layer, an intelligent layer and a database, the intelligent layer comprises a data acquisition module, a data preprocessing module, a concentration performance acquisition module, a process parameter optimization module, an early warning module and a man-machine interaction module; a concentration performance acquisition module and a process parameter optimization module are arranged, so that an improved meta-heuristic optimization algorithm is adopted, a neural network model is introduced, and an optimal process parameter group is acquired based on the concentration performance; a bat algorithm in a meta-heuristic optimization algorithm is adopted, a neural network model is combined in a fitness function, the robustness can be improved, meanwhile, the parallel computing capability is improved, and the bat optimization algorithm has the good global search capability, is not prone to falling into a local optimal solution, has few parameters and is easier to adjust and implement.

Description

Technical Field

[0001] The present invention relates to the technical field of vacuum concentration data analysis, and more particularly to a vacuum concentration data analysis system based on cloud computing. Background Art

[0002] With the continuous advancement of science and technology and the in-depth development of digital transformation, extracting valuable information from data has become the key to improving the competitiveness of enterprises. Vacuum concentration is a common chemical unit operation, widely used in the pharmaceutical, food, and chemical industries. During the vacuum concentration process, a large amount of data is generated, such as parameters such as temperature, pressure, and flow. This data is of great significance for optimizing processes and improving production efficiency. Cloud computing, as an emerging information technology, has the advantages of high-performance computing, distributed storage, elastic expansion, and low cost, providing an effective solution for big data processing.

[0003] However, traditional data analysis methods have problems such as difficulty in data collection, low processing efficiency, and inaccurate analysis results. They cannot meet the real-time and accuracy requirements of modern enterprises for data processing. At the same time, the utilization rate of vacuum concentrated data is low, and vacuum concentrated data cannot be deeply analyzed, thus failing to obtain the hidden data value.

[0004] In view of this, the present invention proposes a vacuum concentration data analysis system based on cloud computing. By applying cloud computing technology to vacuum concentration data analysis, it can realize rapid processing and analysis of large-scale data while mining the hidden data value of vacuum concentration data, providing strong support for enterprise decision-making and laying a data foundation for optimizing processes and improving production efficiency. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a vacuum concentration data analysis system based on cloud computing to solve the problems existing in the above-mentioned background technology.

[0006] The present invention provides the following technical solutions: a vacuum concentration data analysis system based on cloud computing, comprising an equipment layer, a control layer, an intelligent layer, and a database;

[0007] The equipment layer includes acquisition equipment and process equipment;

[0008] The intelligent layer is used to analyze process parameter data, including a data acquisition module, a data preprocessing module, a concentration performance acquisition module, a process parameter optimization module, an early warning module, and a human-computer interaction module;

[0009] The data acquisition module is used to acquire process parameter data and concentrated data of the target object;

[0010] The data preprocessing module is used to perform preprocessing operations on the data;

[0011] The concentration performance acquisition module obtains the concentration effect, concentration quality and concentration energy consumption based on the pre-processed process parameter data and concentration data, and obtains the concentration performance by calculating the concentration contribution and concentration loss; and transmits the data to the process parameter optimization module and the early warning module;

[0012] The process parameter optimization module adopts an improved meta-heuristic optimization algorithm and introduces a neural network model to obtain the optimal process parameter group based on the concentration performance;

[0013] The early warning module is used to receive data from the concentration performance acquisition module and perform early warning judgment, and send an early warning notification to the human-computer interaction module based on the judgment result;

[0014] The human-computer interaction module is used to perform human-computer interaction display on the data;

[0015] The database is used to store historical data and build a knowledge base;

[0016] The control layer is used to receive data from the process parameter optimization module and perform corresponding adjustments based on the values of the optimal process parameter group.

[0017] Preferably, the acquisition device is used to acquire process parameter data, and the acquisition device includes a temperature sensor, a pressure sensor, and a flow sensor; the process parameter data includes pressure, temperature, flow, liquid level, and defoaming frequency of each process link during the vacuum concentration process;

[0018] The target object is the raw material to be vacuum concentrated, and the concentration data is the concentration critical value of the target object; the process parameter data is obtained from the acquisition equipment of the equipment layer, and the concentration data of the target object is obtained from the knowledge base constructed by the database.

[0019] Preferably, the calculation formulas for the concentration effect, concentration quality and concentration energy consumption are respectively expressed as:

[0020] Wherein, NS_xg is the concentration effect, nd_cs is the initial concentration of the target substance in the target object, that is, the concentration of the target substance before vacuum concentration, nd_end is the concentration of the target substance in the target object after vacuum concentration, and ns_t is the total time used for vacuum concentration. The target substance is the substance to be formed after vacuum concentration.

[0021] Among them, NS_z is the concentrated mass, y i _cs is the initial content of the i-th nutrient component of the target substance, y i_end is the content of the i-th nutrient after vacuum concentration of the target substance, Y is the total number of nutrient types, i = 1, 2, 3, ..., Y;

[0022] Among them, NS_nh is the concentration energy consumption, ny j _nh is the total consumption of the jth energy in the vacuum concentration process, J is the total number of energy types, j = 1, 2, 3, ..., J.

[0023] Preferably, the calculation formulas for the concentration contribution and the concentration loss are respectively expressed as:

[0024] The calculation formula of the concentration contribution is expressed as: The calculation formula of the concentration loss is expressed as: Among them, NS_gx is the enrichment contribution and nd_LJ is the enrichment critical value.

[0025] Preferably, the calculation formula of the concentration performance is expressed as:

[0026] Where NS_XN is the concentration performance, k1 and k2 are the corresponding proportional coefficients, k1>k2, and k′ is the concentration effect index. When the concentration contribution is positive, the concentration performance is calculated using formula ①. When the concentration loss is positive, the concentration performance is calculated using formula ②.

[0027] Preferably, the concentration effect index is obtained by setting index demarcation thresholds YUI and YU2; when NS_xg<YU1, the concentration effect index takes a value of 1; when YUI≤NS_xg<YU2, the concentration effect index takes a value of 2; when NS_xg≥YU2, the concentration effect index takes a value of 3; YU1 is the lowest acceptable concentration efficiency value, and YU2 is the highest historical concentration efficiency value.

[0028] Preferably, the specific method for the process parameter optimization module to obtain the optimal process parameter group is:

[0029] Step S11: The process parameter data of a vacuum concentration is regarded as a process parameter group, all process parameter groups are encoded as a, and R process parameter groups are randomly selected to construct the initial population, and all parameters are initialized; the initial iteration number λ is 0; each process parameter group can be regarded as a bat, and the position of the rth bat is p r , the speed is v r , the sound wave frequency is H r , the loudness of the sound wave is B r , the pulse frequency is I r ;

[0030] Step S12: determining a first fitness function;

[0031] Step S13: Find the optimal position p* of the bats in the current population and update the bat position and speed; the optimal position of the bats in the current population is the bat position corresponding to the first maximum fitness value in this iteration;

[0032] Step S14: Generate a random number rand1 between [0, 1]. If rand1>I r , then random flight is performed, and a new position p is generated near the original optimal position by random flight new , otherwise update the bat position according to the bat position update formula;

[0033] Step S15: Generate a random number rand3 on [0, 1]. If rand3 < B r , and the first fitness corresponding to the new optimal position is greater than the first fitness of the original optimal position, then the position is accepted and the new position is used as the current optimal position, and the sound wave loudness is reduced and the pulse frequency is increased;

[0034] Step S16: Determine whether the maximum number of iterations has been reached. If so, the iteration ends, and the process parameter group at the optimal position corresponding to the first maximum fitness value is obtained as the optimal process parameter group; otherwise, set λ=λ+1, and loop steps S13 to S16.

[0035] Preferably, the initial population is represented as: A = {a1, a1, a1, ..., a R}, where A represents the initial population, a r is the rth process parameter group, r = 1, 2, 3, ..., R;

[0036] The formula of the first fitness function in step S12 is expressed as: SY r =NS r _XN′, where SY r is the first fitness value of the process parameter group corresponding to the rth bat, NS r _XN′ is the predicted concentration performance of the process parameter group corresponding to the rth bat;

[0037] The method for obtaining the predicted concentration performance is:

[0038] The process parameter data of the process parameter group corresponding to the r-th bat is input into the concentration performance prediction model to predict the corresponding concentration performance.

[0039] Preferably, the bat position update formula is: Among them, p r λis the position of the rth bat at the λth iteration, p r λ-1 is the position of the rth bat at the λ-1th iteration; v r λ is the speed of the r-th bat at the λ-th iteration;

[0040] The bat speed update formula is: Among them, v r λ-1 is the speed of the rth bat at the λ-1th iteration, H r is the frequency of the sound wave emitted by the rth bat;

[0041] H r =H min +(H max -H min )×β, where H min is the minimum frequency of the bat's sound waves, H max is the maximum frequency of the sound waves generated by bats, and the sound wave frequency range is [H min , H max ], β is a random vector between [0,1].

[0042] Preferably, the formula for the random flight is:

[0043] p new =p old +rand2B λ , where p old is the original optimal position, rand2 is a random number between [0, 1], B λ is the average loudness of all bats in the λth iteration;

[0044] The formula for adjusting the loudness and frequency of sound waves is:

[0045] Among them, α∈(0,1) is the sound wave loudness attenuation coefficient, γ>0 is the pulse frequency enhancement coefficient, B r λ+1 is the sound wave loudness of the rth bat at the λ+1th iteration, B r λ is the sound wave loudness of the r-th bat at the λ-th iteration, I r λ+1 is the pulse frequency of the rth bat at the λ+1th iteration, I r 0 is the initial pulse frequency of the rth bat.

[0046] Technical effects and advantages of the present invention:

[0047] The present invention is provided with a concentration performance acquisition module and a process parameter optimization module, which is beneficial to obtaining the concentration effect, concentration quality and concentration energy consumption based on the pre-processed process parameter data and concentration data, and obtaining the concentration performance by calculating the concentration contribution and concentration loss; adopting the improved meta-heuristic optimization algorithm, introducing the neural network model, and obtaining the optimal process parameter group based on the concentration performance; adopting the bat algorithm in the meta-heuristic optimization algorithm, combining the neural network model in the fitness function, can increase the robustness and improve the parallel computing capability at the same time. The bat optimization algorithm has a good global search capability and is not easy to fall into the local optimal solution. It has fewer parameters and is easier to adjust and implement. During the search process, it can dynamically adjust the search frequency and speed according to the quality of the current solution, so that the algorithm has different search characteristics at different stages, which is helpful to find a better solution. The introduction of the neural network model can give full play to the nonlinear and multimodal data modeling capabilities of the neural network and better capture complex data relationships; at the same time, based on the parallel computing processing mechanism, the computing efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a structural diagram of the vacuum concentration data analysis system based on cloud computing of the present invention.

[0049] Figure 2 This is the intelligent layer flow chart of the present invention. DETAILED DESCRIPTION

[0050] The technical solutions of the present invention will be described clearly and completely below in conjunction with the drawings in the present invention. In addition, the forms of the various structures described in the following embodiments are merely examples. The cloud computing-based vacuum concentration data analysis system involved in the present invention is not limited to the various structures described in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] Example 1

[0052] like Figure 1 As shown, the present invention provides a vacuum concentration data analysis system based on cloud computing, including an equipment layer, a control layer, an intelligent layer and a database;

[0053] The equipment layer includes acquisition equipment and process equipment; the acquisition equipment is used to acquire process parameter data, including but not limited to temperature sensors, pressure sensors, flow sensors, and other devices that acquire process parameter data; the process parameter data include but not limited to pressure, temperature, flow, liquid level, and defoaming frequency of each process link in the vacuum concentration process; each process link corresponds to a type of process equipment; if there are n process links, there are n types of process equipment;

[0054] The control layer includes n control nodes, each of which controls a corresponding process equipment, and the control node is used to adjust the process parameters; the control layer is used to receive data from the process parameter optimization module and perform corresponding adjustments based on the value of the optimal process parameter group;

[0055] The intelligent layer is used to analyze process parameter data, including a data acquisition module, a data preprocessing module, a concentration performance acquisition module, a process parameter optimization module, an early warning module, and a human-computer interaction module;

[0056] The data acquisition module is used to acquire process parameter data and target object concentration data; the target object is the raw material undergoing vacuum concentration, and the concentration data is the target object's concentration threshold. Different items have different concentration thresholds, meaning that when an item reaches a certain concentration value, concentration is considered complete, and this concentration value can be defined as the target object's own concentration threshold. The process parameter data can be acquired from the acquisition equipment at the equipment layer, and the target object concentration data can be acquired from a knowledge base constructed in a database.

[0057] The data preprocessing module is used to perform preprocessing operations on the data to obtain data that can be directly used; the preprocessing operations include but are not limited to cleaning, noise reduction and normalization of the data;

[0058] The concentration performance acquisition module obtains the concentration effect, concentration quality and concentration energy consumption based on the pre-processed process parameter data and concentration data, and obtains the concentration performance by calculating the concentration contribution and concentration loss; and transmits the data to the process parameter optimization module and the early warning module;

[0059] The process parameter optimization module adopts an improved meta-heuristic optimization algorithm, introduces a neural network model, and obtains the optimal process parameter group based on the concentration performance; transmits data to the human-computer interaction module and the control layer; its purpose is to adopt the bat algorithm in the meta-heuristic optimization algorithm, combine the neural network model in the fitness function, and increase robustness while improving parallel computing capabilities. The bat optimization algorithm has good global search capabilities and is not easy to fall into local optimal solutions. It has fewer parameters and is easier to adjust and implement. During the search process, it can dynamically adjust the search frequency and speed according to the quality of the current solution, so that the algorithm has different search characteristics at different stages, which helps to find a better solution. The introduction of the neural network model can give full play to the nonlinear and multimodal data modeling capabilities of the neural network and better capture complex data relationships; at the same time, based on the parallel computing processing mechanism, it improves computing efficiency;

[0060] The early warning module is used to receive data from the concentration performance acquisition module and perform early warning judgment, and send an early warning notification to the human-computer interaction module based on the judgment result;

[0061] The human-computer interaction module is used to perform human-computer interaction display on the data;

[0062] The database is used to store historical data and construct a knowledge base to provide rule knowledge, experience knowledge, common sense knowledge and other contents of each process link of vacuum concentration; the knowledge base can be constructed in the form of a knowledge graph; the knowledge graph construction is an existing technology, and this embodiment will not go into details about it.

[0063] In this embodiment, it should be specifically explained that the specific manner in which the concentration performance acquisition module acquires the concentration performance is:

[0064] Step S01: Calculate the concentration effect, concentration quality, and concentration energy consumption: The calculation formula of the concentration effect is expressed as: Wherein, NS_xg is the concentration effect, nd_cs is the initial concentration of the target substance in the target object, that is, the concentration of the target substance before vacuum concentration, nd_end is the concentration of the target substance in the target object after vacuum concentration, and ns_t is the total time used for vacuum concentration. The target substance is the substance to be formed after vacuum concentration. The better the concentration effect, the higher the concentration efficiency, that is, the shorter the time required to reach the desired concentration concentration, that is, the better the concentration performance;

[0065] The calculation formula of the concentrated mass is expressed as: Among them, NS_z is the concentrated mass, y i _cs is the initial content of the i-th nutrient component of the target substance, y i _end is the content of the i-th nutrient after vacuum concentration of the target substance, Y is the total number of nutrient types, i = 1, 2, 3, ..., Y; the higher the concentration quality, the fewer nutrients lost during the vacuum concentration process, that is, the more nutrients retained, and the better the concentration performance;

[0066] The calculation formula of the concentration energy consumption is expressed as: Among them, NS_nh is the concentration energy consumption, ny j _nh is the total consumption of the jth energy in the vacuum concentration process, J is the total number of energy types, j = 1, 2, 3, ..., J;

[0067] Step S02: Calculate the concentration contribution and the concentration loss: The calculation formula of the concentration contribution is expressed as: The calculation formula of the concentration loss is expressed as: Wherein, NS_gx is the concentration contribution, and nd_LJ is the concentration critical value; the concentration contribution is expressed as the proportion of concentration values exceeding the concentration critical value after the vacuum concentration is completed, that is, the portion contributed within the vacuum concentration concentration; the concentration loss is expressed as the proportion of concentration values that do not reach the concentration critical value after the vacuum concentration is completed, that is, the portion lost in the vacuum concentration concentration; the higher the concentration contribution, the more the vacuum concentration can exceed the concentration critical value, which can be understood as the concentration of the vacuum concentration can exceed its own concentration critical value, that is, the better the vacuum concentration performance; the higher the concentration loss, the more the portion of the vacuum concentration that does not reach the concentration critical value, which can be understood as the concentration of the vacuum concentration cannot reach the concentration critical value, that is, the worse the vacuum concentration performance;

[0068] The step S03: calculating the concentration performance: the calculation formula of the concentration performance is expressed as: Wherein, NS_XN is the concentration performance, k1 and k2 are corresponding proportional coefficients, both of which satisfy the requirement of being greater than zero and less than 1, k1>k2, and k′ is the concentration effect index. When the concentration contribution is positive, the concentration performance is calculated using formula ①. When the concentration loss is positive, the concentration performance is calculated using formula ②. When the concentration contribution and the concentration loss are both zero, formula ① is the same as formula ②, and the simplified formula is used for calculation. The specific values of k1 and k2 can be set by those skilled in the art according to specific circumstances.

[0069] The concentration effect index is obtained by setting index demarcation thresholds YUI and YU2. When NS_xg < YU1, the concentration effect index takes a value of 1; when YUI ≤ NS_xg < YU2, the concentration effect index takes a value of 2; and when NS_xg ≥ YU2, the concentration effect index takes a value of 3. YU1 is the lowest acceptable concentration efficiency, and YU2 is the highest historical concentration efficiency. The specific values of YU1 and YU2 can be set by those skilled in the art based on actual data and actual concentration conditions.

[0070] In this embodiment, it should be specifically explained that the specific manner in which the process parameter optimization module obtains the optimal process parameter group is:

[0071] Step S11: The process parameter data of a vacuum concentration is regarded as a process parameter group, all process parameter groups are encoded as a, and R process parameter groups are randomly selected to construct the initial population, and all parameters are initialized; the initial iteration number λ is 0; each process parameter group can be regarded as a bat, and the position of the rth bat is p r , the speed is v r , the sound wave frequency is H r , the loudness of the sound wave is B r , the pulse frequency is I r,When searching for prey, bats automatically adjust the wavelength and loudness based on the distance between themselves and the target;

[0072] The process parameter data in the process parameter group can be reasonably modified and obtained by technical personnel in the field based on the process parameter data of each historical vacuum concentration, combined with the rule knowledge, experience knowledge and common sense knowledge of each process link in the knowledge base;

[0073] Step S12: determining a first fitness function;

[0074] Step S13: Find the optimal position p* of the bats in the current population and update the bat position and speed; the optimal position of the bats in the current population is the bat position corresponding to the first maximum fitness value in this iteration;

[0075] The bat position update formula is: Among them, p r λ is the position of the rth bat at the λth iteration, p r λ-1 is the position of the rth bat at the λ-1th iteration; v r λ is the speed of the r-th bat at the λ-th iteration;

[0076] The bat speed update formula is: Among them, v r λ-1 is the speed of the rth bat at the λ-1th iteration, H r is the frequency of the sound wave emitted by the rth bat;

[0077] H r =H min +(H max -H min )×β, where H min is the minimum frequency of the bat's sound waves, H max is the maximum frequency of the sound waves generated by bats, and the sound wave frequency range is [H min , H max ], β is a random vector between [0,1];

[0078] Step S14: Generate a random number rand1 between [0, 1]. If rand1>I r , then random flight is performed, and a new position p is generated near the original optimal position by random flight new Otherwise, the bat position is updated according to the bat position update formula; the formula for random flight is:

[0079] p new =p old+rand2B λ , where p old is the original optimal position, rand2 is a random number between [0, 1], B λ is the average loudness of all bats in the λth iteration;

[0080] When searching for prey, bats constantly adjust the loudness and frequency of their sound waves according to the location of their target prey to improve their hunting efficiency. As they gradually approach their prey, the spatial range within which they search for prey decreases. Therefore, they gradually reduce the loudness to a constant value while continuously increasing the frequency in order to quickly and dynamically grasp the location of their target prey.

[0081] Step S15: Generate a random number rand3 on [0, 1]. If rand3 < B r , and the first fitness corresponding to the new optimal position is greater than the first fitness of the original optimal position, then the position is accepted and the new position is used as the current optimal position, and the sound wave loudness is reduced and the pulse frequency is increased;

[0082] The formula for adjusting the loudness and frequency of sound waves is:

[0083] Among them, α∈(0,1) is the sound wave loudness attenuation coefficient, γ>0 is the pulse frequency enhancement coefficient, B r λ+1 is the sound wave loudness of the rth bat at the λ+1th iteration, B r λ is the sound wave loudness of the r-th bat at the λ-th iteration, I r λ+1 is the pulse frequency of the rth bat at the λ+1th iteration, I r 0 is the initial pulse frequency of the rth bat;

[0084] For any sound wave loudness attenuation coefficient and pulse frequency enhancement coefficient, when λ→∞, we have When B r λ When it approaches 0, it can be considered that the bat has found prey and temporarily stops emitting pulses. The pulse variation range can be set by technicians in this field according to specific circumstances. Only when the bat's position is optimized will the loudness and frequency of the pulse be updated, which indicates that the bat is moving towards the optimal position.

[0085] Step S16: Determine whether the maximum number of iterations has been reached. If so, the iteration ends, and the process parameter set corresponding to the optimal position of the first fitness maximum value is obtained as the optimal process parameter set; otherwise, set λ=λ+1, and loop through steps S13 to S16.

[0086] The method of judging whether the maximum number of iterations is reached is: when the value of the first fitness function converges to a certain stable value, it is judged that the maximum number of iterations is reached.

[0087] In this embodiment, it should be specifically explained that the initial population is represented by: A = {a1, a1, a1, ..., a R}, where A represents the initial population, a r is the rth process parameter group, r = 1, 2, 3, ..., R;

[0088] The formula of the first fitness function in step S12 is expressed as: SY r =NS r _XN′, where SY r is the first fitness value of the process parameter group corresponding to the rth bat, NS r _XN′ is the predicted concentration performance of the process parameter group corresponding to the rth bat;

[0089] The method for obtaining the predicted concentration performance is:

[0090] The process parameter data of the process parameter group corresponding to the rth bat is input into the concentration performance prediction model to predict the corresponding concentration performance; the training process of the concentration performance prediction model is as follows:

[0091] Pre-collecting d sets of process parameter data as analysis data, obtaining the concentration performance corresponding to the d sets of process parameter data, and converting a set of process parameter data and the corresponding concentration performance into a set of corresponding feature vectors, where d is an integer greater than 1, i.e., there are d sets of feature vectors;

[0092] The d groups of feature vectors are used as the input of the concentration performance prediction model, the predicted concentration performance is used as the output of the concentration performance prediction model, and the actual concentration performance corresponding to each group of process parameter data is used as the prediction target. The actual concentration performance can be obtained by calculating the formula in step S03; the training goal is to minimize the sum of the prediction errors of all analysis data; the formula for the prediction error is expressed as: ε p =θ p -μ p , where ε p is the prediction error, p is the group number of the eigenvector corresponding to the analysis data, θ p is the predicted concentration performance corresponding to the p-th group of analysis data, μ p The concentration performance prediction model is trained for the actual concentration performance corresponding to the p-th group of analysis data, and the training is stopped when the sum of the prediction errors reaches convergence; the concentration performance prediction model is specifically a neural network model.

[0093] In this embodiment, it should be specifically explained that the specific manner in which the early warning module performs early warning determination is:

[0094] If the concentration performance satisfies NS_XN≥UI, the judgment result is that no warning is required; at this time, the concentration performance meets the set threshold, which means that the concentration performance meets the requirements and no warning prompt is required, and vacuum concentration can be carried out normally;

[0095] If the concentration performance satisfies NS_XN<UI, the judgment result is that an early warning is required; at this time, the concentration performance does not reach the set threshold, which means that the concentration performance does not meet the requirements, and an early warning prompt is required to review the vacuum concentration process, including but not limited to checking the equipment or improving the concentration effect and quality;

[0096] The UI is a warning threshold, which satisfies UI≥50%. The specific value of the warning threshold can be set by those skilled in the art according to actual conditions.

[0097] Example 2

[0098] The present invention provides a vacuum concentration data analysis system based on cloud computing, comprising an equipment layer, a control layer, an intelligent layer and a database;

[0099] The equipment layer includes acquisition equipment and process equipment; the acquisition equipment is used to acquire process parameter data, including but not limited to temperature sensors, pressure sensors, flow sensors, and other devices that acquire process parameter data; the process parameter data include but not limited to pressure, temperature, flow, liquid level, and defoaming frequency of each process link in the vacuum concentration process; each process link corresponds to a type of process equipment; if there are n process links, there are n types of process equipment;

[0100] The control layer includes n control nodes, each of which controls a corresponding process equipment, and the control node is used to adjust the process parameters;

[0101] like Figure 2 As shown, the intelligent layer is used to analyze process parameter data, including a data acquisition module, a data preprocessing module, a concentration performance acquisition module, a process parameter optimization module, an early warning module and a human-computer interaction module;

[0102] The data acquisition module is used to acquire process parameter data and concentration data of a target object; the target object is the raw material to be vacuum concentrated, and the concentration data is the concentration critical value of the target object;

[0103] The data preprocessing module is used to perform preprocessing operations on the data to obtain data that can be directly used;

[0104] The concentration performance acquisition module obtains the concentration effect, concentration quality and concentration energy consumption based on the pre-processed process parameter data and concentration data, and obtains the concentration performance by calculating the concentration contribution and concentration loss; and transmits the data to the process parameter optimization module and the early warning module;

[0105] The process parameter optimization module adopts an improved meta-heuristic optimization algorithm, introduces a neural network model, and sets an objective function to obtain the optimal process parameter group;

[0106] The early warning module is used to receive data from the concentration performance acquisition module and perform early warning judgment, and send an early warning notification to the human-computer interaction module based on the judgment result;

[0107] The human-computer interaction module is used to perform human-computer interaction display on the data;

[0108] The database is used to store data and build a knowledge base. The knowledge base can be built in the form of a knowledge graph.

[0109] In this embodiment, it should be specifically explained that the specific manner in which the process parameter optimization module obtains the optimal process parameter group is:

[0110] Step S21: The process parameter data of a vacuum concentration is regarded as a process parameter group, all process parameter groups are encoded as a, and R process parameter groups are randomly selected to construct the initial population, and all parameters are initialized; the initial iteration number λ is 0; each process parameter group can be regarded as a bat, and the position of the rth bat is p r , the speed is v r , the sound wave frequency is H r , the loudness of the sound wave is B r , the pulse frequency is I r ,When searching for prey, bats automatically adjust the wavelength and loudness based on the distance between themselves and the target;

[0111] Step S22: determining a second fitness function;

[0112] Step S23: Find the optimal position p* of the bats in the current population and update the bat position and speed; the optimal position of the bats in the current population is the bat position corresponding to the second maximum fitness value in this iteration;

[0113] Step S24: Generate a random number rand1 between [0, 1]. If rand1>I r , then random flight is performed, and a new position p is generated near the original optimal position by random flight new , otherwise update the bat position according to the bat position update formula;

[0114] Step S25: Generate a random number rand3 on [0, 1]. If rand3 < B r , and the second fitness corresponding to the new optimal position is greater than the second fitness of the original optimal position, then the position is accepted and the new position is used as the current optimal position, and the sound wave loudness is reduced and the pulse frequency is increased;

[0115] Step S26: Determine whether the maximum number of iterations has been reached. If so, the iteration ends, and the process parameter set corresponding to the optimal position of the second maximum fitness value is obtained as the optimal process parameter set; otherwise, set λ=λ+1, and loop through steps S23 to S26.

[0116] The method of determining whether the maximum number of iterations has been reached is: when the value of the second fitness function converges to a certain stable value, it is determined that the maximum number of iterations has been reached.

[0117] In this embodiment, it should be specifically explained that the formula of the second fitness function in step S22 is expressed as: SY r ′=MB r _m, where SY r ′ is the second fitness value of the process parameter group corresponding to the rth bat, MB r _m is the maximum or minimum value of the objective function of the process parameter group corresponding to the rth bat;

[0118] The objective function may be any one of the concentration effect, concentration quality, and concentration energy consumption, or other functions such as concentration cost, and may be set by those skilled in the art. The maximum or minimum value may be selected according to different objective functions. Alternatively, a multi-objective function may be combined to form a second fitness function. The specific form of the objective function may be flexibly selected by those skilled in the art.

[0119] Other details not described in this embodiment are the same as those in Example 1.

[0120] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0121] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A cloud computing-based vacuum concentration data analysis system, characterized by: Including equipment layer, control layer, intelligent layer and database; The equipment layer includes acquisition equipment and process equipment; The intelligent layer is used to analyze process parameter data, including a data acquisition module, a data preprocessing module, a concentration performance acquisition module, a process parameter optimization module, an early warning module, and a human-computer interaction module; The data acquisition module is used to acquire process parameter data and concentrated data of the target object; The data preprocessing module is used to perform preprocessing operations on the data; The concentration performance acquisition module obtains the concentration effect, concentration quality and concentration energy consumption based on the pre-processed process parameter data and concentration data, and obtains the concentration performance by calculating the concentration contribution and concentration loss; And transmit the data to the process parameter optimization module and early warning module; The process parameter optimization module adopts an improved meta-heuristic optimization algorithm and introduces a neural network model to obtain the optimal process parameter group based on the concentration performance; The early warning module is used to receive data from the concentration performance acquisition module and perform early warning judgment, and send an early warning notification to the human-computer interaction module based on the judgment result; The human-computer interaction module is used to perform human-computer interaction display on the data; The database is used to store historical data and build a knowledge base; The control layer is used to receive data from the process parameter optimization module and perform corresponding adjustments based on the values of the optimal process parameter group.

2. The cloud computing-based vacuum concentration data analysis system according to claim 1, characterized in that: The acquisition equipment is used to collect process parameter data, and the acquisition equipment includes a temperature sensor, a pressure sensor and a flow sensor; the process parameter data includes the pressure, temperature, flow, liquid level and defoaming frequency of each process link in the vacuum concentration process; The target object is the raw material to be vacuum concentrated, and the concentration data is the concentration critical value of the target object; The process parameter data is obtained from the acquisition equipment of the equipment layer, and the concentrated data of the target object is obtained from the knowledge base constructed by the database.

3. The cloud computing-based vacuum concentration data analysis system according to claim 2, characterized in that: The calculation formulas for the concentration effect, concentration quality and concentration energy consumption are respectively expressed as: Wherein, NS_xg is the concentration effect, nd_cs is the initial concentration of the target substance in the target object, that is, the concentration of the target substance before vacuum concentration, nd_end is the concentration of the target substance in the target object after vacuum concentration, and ns_t is the total time used for vacuum concentration. The target substance is the substance to be formed after vacuum concentration. Among them, NS_z is the concentrated mass, y i _cs is the initial content of the i-th nutrient component of the target substance, y i _end is the content of the i-th nutrient after vacuum concentration of the target substance, Y is the total number of nutrient types, i = 1, 2, 3, ..., Y; Among them, NS_nh is the concentration energy consumption, ny j _nh is the total consumption of the jth energy in the vacuum concentration process, J is the total number of energy types, j = 1, 2, 3, ..., J.

4. The cloud computing-based vacuum concentration data analysis system according to claim 3, characterized in that: The calculation formulas for the concentration contribution and the concentration loss are respectively expressed as: The calculation formula of the concentration contribution is expressed as: The calculation formula of the concentration loss is expressed as: Among them, NS_gx is the enrichment contribution and nd_LJ is the enrichment critical value.

5. The cloud computing-based vacuum concentration data analysis system according to claim 4, characterized in that: The calculation formula of the concentration performance is expressed as: Where NS_XN is the concentration performance, k1 and k2 are the corresponding proportional coefficients, k1>k2, and k′ is the concentration effect index. When the concentration contribution is positive, the concentration performance is calculated using formula ①. When the concentration loss is positive, the concentration performance is calculated using formula ②.

6. The cloud computing-based vacuum concentration data analysis system according to claim 5, characterized in that: The concentration effect index is obtained by setting index demarcation thresholds YUI and YU2. When NS_xg is less than YU1, the concentration effect index takes a value of 1; when YUI≤NS_xg<YU2, the concentration effect index takes a value of 2; when NS_xg≥YU2, the concentration effect index takes a value of 3. YU1 is the lowest acceptable concentration efficiency value, and YU2 is the highest historical concentration efficiency value.

7. The cloud computing-based vacuum concentration data analysis system according to claim 6, characterized in that: The specific method for the process parameter optimization module to obtain the optimal process parameter group is: Step S11: The process parameter data of a vacuum concentration is regarded as a process parameter group, all process parameter groups are encoded as a, and R process parameter groups are randomly selected to construct the initial population, and all parameters are initialized; the initial iteration number λ is 0; each process parameter group can be regarded as a bat, and the position of the rth bat is p r , the speed is v r , the sound wave frequency is H r , the loudness of the sound wave is B r , the pulse frequency is I r ; Step S12: determining a first fitness function; Step S13: Find the optimal position p* of the bats in the current population and update the bat position and speed; the optimal position of the bats in the current population is the bat position corresponding to the first maximum fitness value in this iteration; Step S14: Generate a random number rand1 between [0, 1]. If rand1>I r , then random flight is performed, and a new position p is generated near the original optimal position by random flight new , otherwise update the bat position according to the bat position update formula; Step S15: Generate a random number rand3 on [0, 1]. If rand3 < B r , and the first fitness corresponding to the new optimal position is greater than the first fitness of the original optimal position, then the position is accepted and the new position is used as the current optimal position, and the sound wave loudness is reduced and the pulse frequency is increased; Step S16: determining whether the maximum number of iterations has been reached, and if so, the iteration ends, and obtaining the process parameter set at the optimal position corresponding to the first maximum fitness value as the optimal process parameter set; Otherwise, let λ=λ+1 and loop through steps S13 to S16.

8. The cloud computing-based vacuum concentration data analysis system according to claim 7, characterized in that: The initial population is represented as: A = {a1, a1, a1, ..., a R }, where A represents the initial population, a r is the rth process parameter group, r = 1, 2, 3, ..., R; The formula of the first fitness function in step S12 is expressed as: SY r =NS r _XN′, where SY r is the first fitness value of the process parameter group corresponding to the rth bat, NS r _XN′ is the predicted concentration performance of the process parameter group corresponding to the rth bat; The method for obtaining the predicted concentration performance is: The process parameter data of the process parameter group corresponding to the r-th bat is input into the concentration performance prediction model to predict the corresponding concentration performance.

9. The cloud computing-based vacuum concentration data analysis system according to claim 8, characterized in that: The bat position update formula is: Among them, p r λ is the position of the rth bat at the λth iteration, p r λ-1 is the position of the rth bat at the λ-1th iteration; v r λ is the speed of the r-th bat at the λ-th iteration; The bat speed update formula is: Among them, v r λ-1 is the speed of the rth bat at the λ-1th iteration, H r is the frequency of the sound wave emitted by the rth bat; H r =H min +(H max -H min )×β, where H min is the minimum frequency of the bat's sound waves, H max is the maximum frequency of the sound waves generated by bats, and the sound wave frequency range is [H min , H max ], β is a random vector between [0,1].

10. The cloud computing-based vacuum concentration data analysis system according to claim 9, characterized in that: The formula for the random flight is: p new =p old +rand2B λ , where p old is the original optimal position, rand2 is a random number between [0, 1], B λ is the average loudness of all bats in the λth iteration; The formula for adjusting the loudness and frequency of sound waves is: Among them, α∈(0,1) is the sound wave loudness attenuation coefficient, γ>0 is the pulse frequency enhancement coefficient, B r λ+1 is the sound wave loudness of the rth bat at the λ+1th iteration, B r λ is the sound wave loudness of the r-th bat at the λ-th iteration, I r λ+1 is the pulse frequency of the rth bat at the λ+1th iteration, I r 0 is the initial pulse frequency of the rth bat.

Images