Remote intelligent control method and system for concrete production equipment

Abstract

The application provides a remote intelligent control method and system for concrete production equipment, relates to the technical field of concrete production, and comprises the following steps: collecting production parameter data; establishing an equipment correlation topological graph to determine a weight coefficient; dividing dynamic collaboration groups based on a collaboration efficiency index; distributing tasks according to a production capacity index; and generating an optimal instruction sequence by optimizing control parameters through an ant colony algorithm. The application realizes efficient collaborative work and intelligent management of the concrete production equipment, and improves production efficiency and resource utilization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of concrete production, and particularly relates to a remote intelligent control method and system for concrete production equipment. BACKGROUND

[0002] The existing remote control and management system for concrete production equipment has the following defects and deficiencies: the existing system lacks in-depth analysis and utilization of the correlation between concrete production equipment, and each device often operates independently, which cannot realize dynamic collaborative grouping based on correlation topology and weight coefficients, resulting in low collaborative efficiency between devices, unreasonable resource allocation, and difficulty in coping with complex and variable production environments; the existing technology lacks a scientific and effective evaluation mechanism in terms of production task allocation, and usually adopts a static or empirical task allocation method, which fails to dynamically optimize allocation according to the actual production capacity indicators of each device group, resulting in the situation that some devices are overloaded while others are idle, and the overall production efficiency is limited; the existing concrete production control system has single technical means in terms of control strategy optimization, and most of them adopt fixed parameter control or simple PID control, lack adaptive control strategy optimization mechanism based on multi-objective evaluation and intelligent algorithm, and cannot realize dynamic optimization of control parameters, which cannot automatically adjust the control strategy according to the actual production situation, thereby affecting the production quality and efficiency of concrete. SUMMARY

[0003] The embodiments of the present application provide a remote intelligent control method and system for concrete production equipment, which can solve the problems in the prior art.

[0004] In a first aspect, the embodiments of the present application provide a remote intelligent control method for concrete production equipment, comprising:

[0005] Collecting production parameter data of the concrete production equipment; establishing a correlation topology graph between the concrete production equipment according to the production parameter data, and determining correlation weight coefficients; calculating a collaborative efficiency indicator of a device group based on the correlation weight coefficients, and dividing the concrete production equipment into a plurality of dynamic collaborative groups according to the collaborative efficiency indicator;

[0006] Obtaining a concrete production task instruction, and calculating a production capacity indicator of each dynamic collaborative group based on the device distribution of the dynamic collaborative group;

[0007] Allocating a production task to each dynamic collaborative group according to the production capacity indicator, calculating a multi-objective evaluation score based on the completion, constructing a target function of an ant colony algorithm based on the multi-objective evaluation score, taking the weight coefficients of the control parameters as decision variables for iterative optimization, selecting an optimal weight coefficient combination obtained after iterative optimization, and generating an optimal control instruction sequence.

[0008] establishing a correlation topology graph between the concrete production devices according to the production parameter data, determining a correlation weight coefficient, comprising:

[0009] calculating mutual information entropy values under different time delay orders according to the production parameter data, composing a multi-scale mutual information entropy sequence from the mutual information entropy values, and extracting a local maximum point from the multi-scale mutual information entropy sequence as an optimal time delay order;

[0010] statistically analyzing state transition frequencies of the concrete production devices in different running states, and determining a state transition synchronization degree of the concrete production devices according to a difference value of elements in a state transition probability matrix corresponding to the state transition frequencies;

[0011] taking the concrete production devices as nodes, and taking the optimal time delay order and the state transition synchronization degree as weights of edges to generate a correlation topology graph;

[0012] calculating prediction errors of the optimal time delay order and the state transition synchronization degree, determining a weight value according to the prediction errors combined with an exponential function, and determining a correlation weight coefficient according to the optimal time delay order and the state transition synchronization degree and the weight value.

[0013] calculating a coordination efficiency index of a device group based on the correlation weight coefficient, and dividing the concrete production devices into a plurality of dynamic coordination groups according to the coordination efficiency index, comprising:

[0014] adding and normalizing the correlation weight coefficients of each concrete production device and other concrete production devices to obtain a coordination contribution degree representing each concrete production device, and determining a coordination efficiency index according to the correlation weight coefficients of any two concrete production devices in a group and the coordination contribution degrees of all the concrete production devices;

[0015] dividing the concrete production devices with the coordination efficiency index greater than a preset screening threshold into a group to obtain a candidate grouping scheme, calculating a ratio of an intersection element number and a union element number of coordination group member sets at adjacent two time points, adding the group stability index at a previous time point to obtain a group stability index at a current time point, and dividing the concrete production devices into a plurality of dynamic coordination groups according to the candidate grouping scheme when the group stability index at continuous multiple time points meets a preset stability threshold.

[0016] obtaining a concrete production task instruction, and calculating a production capacity index of each dynamic coordination group based on a device distribution of the dynamic coordination group, comprising:

[0017] Obtaining real-time spatial coordinates of each concrete production equipment in the dynamic synergy group, calculating a spatial distribution center of the dynamic synergy group based on the real-time spatial coordinates, and taking an average distance from each concrete production equipment to the spatial distribution center as a spatial distribution index;

[0018] Matching the production parameter data with the production task instructions to obtain an inter-device matching degree;

[0019] Determining a synergy gain coefficient according to the inter-device matching degree and the spatial distribution index in the dynamic synergy group, multiplying a basic production capacity of the concrete production equipment by a weighted function value of the synergy gain coefficient to obtain a production capacity index of each dynamic synergy group.

[0020] Based on the multi-objective evaluation score, constructing a target function of an ant colony algorithm, taking a weight coefficient of a control parameter as a decision variable for iterative optimization, selecting an optimal weight coefficient combination obtained after the iterative optimization to generate an optimal control instruction sequence, including:

[0021] Initializing a weight coefficient of an ant colony individual and constructing a target function of an ant colony algorithm based on the multi-objective evaluation score, recording a local optimal solution and a global optimal solution obtained by the ant colony individual in the current iteration based on the target function, adding the local optimal solution and the global optimal solution after being multiplied by a preset reinforcement coefficient to obtain a current path pheromone concentration;

[0022] Constructing a candidate path set according to the current path pheromone concentration, calculating a dynamic heuristic factor that is adaptively adjusted in the iteration process based on the current iteration progress, obtaining a selection probability of quantifying the path attraction degree according to the current path pheromone concentration multiplied by the dynamic heuristic factor and divided by a sum of products of pheromone concentrations of all paths in the candidate path set;

[0023] Based on the selection probability, sampling a path and calculating an evaluation score of the current path, calculating a parameter step size according to the current iteration number, and calculating a weight coefficient of the current path according to the parameter step size, the evaluation score of the current path, and a historical optimal evaluation score;

[0024] Generating an optimal control instruction sequence based on the weight coefficient and a preset basic control amount.

[0025] Calculating a dynamic heuristic factor that is adaptively adjusted in the iteration process based on the current iteration progress, including:

[0026] Substituting the current iteration progress into a sinusoidal fluctuation function to generate a fluctuation period factor;

[0027] The optimal objective function value obtained according to the current iteration and the historical optimal objective function value are used to determine a convergence state, the convergence state and the fluctuation period factor are combined into a cosine decreasing function to obtain a stage attenuation parameter, a search intensity parameter is determined according to the objective function values of all paths in the current iteration population, and a convergence evaluation value is generated according to the attenuation parameter and the search intensity parameter;

[0028] Manhattan distance calculation is performed on the paths in the candidate path set, a relative distance parameter is obtained based on distance distribution characteristics, the relative distance parameter and the convergence evaluation value are nonlinearly mapped through an S-shaped function to obtain a diversity adjustment parameter;

[0029] A dynamic heuristic factor is generated in combination with the convergence evaluation value and the diversity adjustment parameter.

[0030] The method further comprises:

[0031] Local running data is collected and trained to obtain local optimization parameters, the local optimization parameters are encrypted through a differential privacy encryption algorithm, the encrypted local optimization parameters are uploaded to the cloud end for aggregation operation to generate global optimization parameters, and production tasks are allocated to each dynamic cooperative group according to the global optimization parameters and the production capacity indicators.

[0032] In a second aspect of the embodiment of the application, a remote intelligent control system for a concrete production device is provided, comprising:

[0033] A first unit is configured to collect production parameter data of the concrete production device, establish an associated topology graph among the concrete production devices according to the production parameter data, and determine an associated weight coefficient; the cooperative efficiency indicators of device groups are calculated based on the associated weight coefficient, and the concrete production devices are divided into a plurality of dynamic cooperative groups according to the cooperative efficiency indicators;

[0034] A second unit is configured to obtain a concrete production task instruction, and calculate production capacity indicators of each dynamic cooperative group based on the distribution of devices in the dynamic cooperative groups;

[0035] A third unit is configured to allocate production tasks to each dynamic cooperative group according to the production capacity indicators, calculate a multi-objective evaluation score based on the completion, construct a target function of an ant colony algorithm based on the multi-objective evaluation score, use the weight coefficients of control parameters as decision variables to perform iterative optimization, select an optimal weight coefficient combination obtained after the iterative optimization, and generate an optimal control instruction sequence.

[0036] In a third aspect of the embodiment of the application,

[0037] An electronic device is provided, comprising:

[0038] A processor;

[0039] a memory for storing processor-executable instructions;

[0040] wherein the processor is configured to invoke the instructions stored by the memory to perform the aforementioned method.

[0041] A fourth aspect of the embodiments of the present application,

[0042] A computer-readable storage medium is provided, which stores computer program instructions, and the computer program instructions are executed by a processor to implement the aforementioned method.

[0043] The beneficial effects of the present application are as follows:

[0044] The present application divides the concrete production equipment into multiple dynamic collaborative groups and calculates the collaborative efficiency index of the equipment group based on the correlation weight coefficient, realizing intelligent grouping and collaborative control of the concrete production equipment, and significantly improving the overall collaborative efficiency and resource utilization of the production process.

[0045] By constructing an objective function based on the ant colony algorithm and performing iterative optimization, the present application can generate an optimal control instruction sequence, making the concrete production process more accurate and controllable, reducing production deviation, improving the quality stability and production consistency of the concrete, and reducing energy consumption and raw material waste.

[0046] The present application constructs a distributed intelligent control system for concrete production based on production parameter data, realizes remote real-time monitoring and intelligent management of production equipment, improves system response speed and data processing efficiency, and enhances the reliability and stability of the production system. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 A flowchart of the remote intelligent control method for the concrete production equipment of the embodiments of the present application is shown.

[0048] Figure 2 A flowchart of the correlation weight coefficient calculation is shown. DETAILED DESCRIPTION

[0049] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0050] The technical solutions of the present application are described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes may not be described in detail in some examples.

[0051] Figure 1 The flowchart of the remote intelligent control method for the concrete production equipment of the embodiments of the present application is shown in FIG. 1. Figure 1 The method comprises the following steps:

[0052] Collecting production parameter data of the concrete production equipment; establishing an associated topology graph between the concrete production equipment according to the production parameter data, determining an associated weight coefficient; calculating a coordination efficiency index of the equipment group based on the associated weight coefficient, and dividing the concrete production equipment into multiple dynamic coordination groups according to the coordination efficiency index;

[0053] Obtaining a concrete production task instruction, calculating a production capacity index of each dynamic coordination group based on the distribution of the equipment of the dynamic coordination group;

[0054] Allocating a production task to each dynamic coordination group according to the production capacity index, calculating a multi-objective evaluation score based on the completion, constructing a target function of an ant colony algorithm based on the multi-objective evaluation score, taking the weight coefficient of the control parameter as a decision variable for iterative optimization, selecting an optimal weight coefficient combination obtained after the iterative optimization, and generating an optimal control instruction sequence.

[0055] Figure 2 The flowchart for calculating the associated weight coefficient is shown in FIG. 2. In an optional embodiment, the associated topology graph between the concrete production equipment is established according to the production parameter data, and the associated weight coefficient is determined, comprising the following steps:

[0056] Calculating mutual information entropy values under different time delay orders according to the production parameter data, composing a multi-scale mutual information entropy sequence from the mutual information entropy values, and extracting a local maximum point from the multi-scale mutual information entropy sequence as an optimal time delay order;

[0057] Statistically analyzing the state transition frequencies of the concrete production equipment in different operating states, and determining the state transition synchronization degree of the concrete production equipment according to the difference of the elements in the state transition probability matrix corresponding to the state transition frequencies;

[0058] Taking the concrete production equipment as nodes, and taking the optimal time delay order and the state transition synchronization degree as the weight of the edges to generate an associated topology graph;

[0059] The prediction error of the optimal time delay order and the state transition synchronization degree is calculated, a weight value is determined according to the prediction error combined with an exponential function, and an association weight coefficient is determined according to the optimal time delay order and the state transition synchronization degree and the weight value.

[0060] In a concrete production line, production parameter data of multiple devices are collected, including the running state, temperature, speed, material flow and other parameters of devices such as mixers, metering systems, conveyors and the like. These parameter data are recorded in time series form, with a sampling frequency of one per second, continuously collected for 72 hours, generating a total of 259200 data points.

[0061] For calculating the mutual information entropy value under different time delay orders, the parameter data of each device is first standardized to have a mean value of 0 and a variance of 1. Then, the time delay order is set to a value range of 0 to 100, i.e. a delay of 0 to 100 seconds. For each delay order τ, the mutual information entropy value between device 1 and device 2 is calculated. Specifically, the parameter data sequence of device 1 is paired with the parameter data sequence of device 2 delayed by τ time, the joint distribution probability and the marginal distribution probability are counted, and then the mutual information entropy value is calculated. Taking the mixer and the metering system as an example, when the delay order τ = 0, the mutual information entropy value is 0.15; when τ = 5, the mutual information entropy value rises to 0.42; when τ = 27, it reaches a peak of 0.87; and when τ = 60, it drops to 0.23. Arranging these mutual information entropy values according to the time delay order forms a multi-scale mutual information entropy sequence [0.15, 0.21, 0.33, 0.38, 0.42,..., 0.87,..., 0.23, 0.19]. Through a local maximum value detection algorithm, the peak points in the sequence are identified, and in this example, τ = 27 is the local maximum value point, so the optimal time delay order between the mixer and the metering system is determined to be 27 seconds.

[0062] The running state category division of the production parameter data is completed based on cluster analysis. The parameter data of each device is feature extracted, and the mean, standard deviation, peak factor, skewness and kurtosis are selected as features, and the K-means clustering algorithm is used to divide the running state into 4 categories: normal operation, low load, high load and abnormal state. For the mixer, state 1 indicates that the mixing speed is 60-70 rpm and the material is mixed evenly; state 2 indicates that the speed is 45-60 rpm and the material is not mixed well; state 3 indicates that the speed is 70-85 rpm and the material is mixed too much; and state 4 indicates that the speed is abnormally fluctuating or in a shutdown state.

[0063] Based on the divided operating state categories, the state transition frequencies of the concrete production equipment are counted. A state transition matrix is constructed, and the matrix elements represent the number of times of transition from state i to state j. Taking the mixer as an example, the number of times of transition from state 1 to state 1 is 8560, the number of times of transition from state 1 to state 2 is 1240, the number of times of transition from state 1 to state 3 is 320, and the number of times of transition from state 1 to state 4 is 80. The transition matrix is normalized to obtain a state transition probability matrix, such as the transition probability from state 1 to state 1 is 0.84, the transition probability from state 1 to state 2 is 0.12, the transition probability from state 1 to state 3 is 0.03, and the transition probability from state 1 to state 4 is 0.01.

[0064] The element difference of the state transition probability matrix between different equipment is calculated. For the mixer and the metering system, the transition probability difference from state 1 to state 1 is |0.84-0.81|=0.03, the transition probability difference from state 1 to state 2 is |0.12-0.13|=0.01, and so on. The average of these differences in all state categories is 0.025 for the mixer and the metering system. The average difference of all equipment pairs is normalized to obtain the state transition synchronization degree. The state transition synchronization degree of the mixer and the metering system is 0.86, indicating that the operating state changes of the two equipment are highly synchronized.

[0065] The concrete production equipment is taken as a node to construct a correlation topology graph. Taking the mixer, the metering system, the conveyor belt, the cement bin, and the sandstone bin as nodes, a total of 10 edges of an undirected graph are formed. Each edge has two weights: the optimal time delay order and the state transition synchronization degree. For example, the edge weight between the mixer and the metering system is (27, 0.86), the edge weight between the mixer and the conveyor belt is (15, 0.72), and the edge weight between the metering system and the conveyor belt is (8, 0.65).

[0066] In order to integrate the two weights, the prediction errors of the optimal time delay order and the state transition synchronization degree are calculated. Through ten-fold cross-validation, the time delay prediction mean square error of the mixer and the metering system is 0.043, and the state transition synchronization degree prediction mean square error is 0.028. The prediction errors are substituted into the exponential function e -误差 , and the time sequence correlation weight coefficient is e -0.043 =0.958, and the state correlation weight coefficient is e -0.028 =0.972.

[0067] Finally, the correlation weight coefficient is calculated by weighted summation. The correlation weight coefficient of the mixer and the metering system is 27 x 0.958 + 0.86 x 0.972 = 26.71. Similarly, the correlation weight coefficients of other pairs of devices are calculated to generate a complete correlation topology graph, thereby realizing quantitative characterization of the relationships between the concrete production devices and providing data support for production scheduling and fault diagnosis.

[0068] In an alternative embodiment, a synergy efficiency indicator of a device group is calculated based on the correlation weight coefficient, and the concrete production devices are divided into a plurality of dynamic synergy groups according to the synergy efficiency indicator, comprising:

[0069] The correlation weight coefficients of each concrete production device and other concrete production devices are added and normalized to obtain a synergy contribution degree representing each concrete production device; and a synergy efficiency indicator is determined according to the correlation weight coefficients of any two concrete production devices in the group and the synergy contribution degrees of all concrete production devices;

[0070] The concrete production devices with the synergy efficiency indicator greater than a preset screening threshold are divided into a group to obtain a candidate grouping scheme, and a ratio of the number of intersection elements to the number of union elements of the synergy group member set at two adjacent time points in the candidate grouping scheme is calculated, and the group stability indicator at the previous time point is added to obtain the group stability indicator at the current time point; when the group stability indicator at consecutive multiple time points all satisfies a preset stability threshold, the concrete production devices are divided into a plurality of dynamic synergy groups in accordance with the candidate grouping scheme.

[0071] In an actual application scenario, first, the running state data of each device in the concrete production system is obtained. Taking a certain concrete production plant as an example, the plant has 8 main production devices, including 2 mixers (labeled as M1 and M2), 3 delivery pumps (labeled as P1, P2, and P3), and 3 loaders (labeled as L1, L2, and L3). The running state data of the devices is collected every 10 minutes, including the key parameters such as the on-time, off-time, running power, and material processing capacity of the devices.

[0072] Based on the collected device running state data, the correlation weight coefficients between the devices are calculated. The correlation weight coefficient reflects the strength of the cooperative operation relationship between two devices. In this embodiment, the correlation weight coefficient is determined by calculating the weighted values of three dimensions, i.e., the device running time overlap, the material flow relationship, and the energy consumption correlation. For example, when the running time of the mixer M1 and the delivery pump P1 is highly overlapped, and the material flow is close, the correlation weight coefficient of the two is high, specifically 0.82; while the synergy relationship between M1 and L3 is weak, and the correlation weight coefficient is only 0.21.

[0073] Next, the collaborative contribution of each device is calculated based on the correlation weight coefficients. For each concrete production device, the correlation weight coefficients with all other devices are added and then normalized to obtain the collaborative contribution of the device. For example, the correlation weight coefficients of M1 with the other 7 devices (0.76, 0.82, 0.45, 0.35, 0.21, 0.33, and 0.28, respectively) are added to obtain 3.2, and the collaborative contribution of M1 is 0.18 after normalization. Similarly, the collaborative contributions of the other devices are calculated: M2 is 0.16, P1 is 0.21, P2 is 0.19, P3 is 0.11, L1 is 0.06, L2 is 0.05, and L3 is 0.04.

[0074] Based on the collaborative contribution of the devices and the correlation weight coefficients, the collaborative efficiency index of the device group is calculated. After selecting a device combination, the correlation weight coefficients of any two devices in the group are multiplied by the corresponding collaborative contribution of the devices and added up. For example, considering the device combination {M1, M2, P1, P2}, the correlation weight coefficient of M1 and M2, 0.76, is multiplied by the product of the collaborative contribution of M1, 0.18, and the collaborative contribution of M2, 0.16, to obtain 0.022. Similarly, the weighted values of all device pairs in the combination are calculated and added up to obtain the original value of the collaborative efficiency of the combination, which is 0.153. This value is divided by the maximum possible value of the sum of the squares of the collaborative contributions of all devices in the group, 0.172, to finally obtain the collaborative efficiency index, which is 0.89.

[0075] According to the calculated collaborative efficiency index, high-efficiency collaborative device combinations are selected. Set the preset screening threshold to 0.75, and for device combinations with a collaborative efficiency index greater than 0.75, they are selected as candidate grouping schemes. In this example, the device combination {M1, M2, P1, P2} has a collaborative efficiency index of 0.89, which is greater than the threshold of 0.75, so it is selected as candidate grouping scheme one; the device combination {P3, L1, L2} has a collaborative efficiency index of 0.81, which is also selected as candidate grouping scheme two; and the device combination {L3} is a single group, forming candidate grouping scheme three.

[0076] To ensure the stability of the grouping scheme, the group stability index needs to be calculated. At consecutive multiple time points, the ratio of the number of intersection elements to the number of union elements of the collaborative group member set at adjacent two time points is calculated. For example, at time t and t+1, the device combinations in candidate grouping scheme one are {M1, M2, P1, P2} and {M1, M2, P1, P2, L1}, respectively, the number of intersection elements is 4, the number of union elements is 5, and the ratio is 0.8. Assuming that the group stability index at the previous time is 0.85, then the group stability index at the current time is 0.8+0.85=1.65.

[0077] A preset stability threshold is set to 1.5, and when the group stability indicators of 5 consecutive time points are all greater than 1.5, it is considered that the grouping scheme is stable and reliable. In this example, after continuous monitoring, the group stability indicators of the candidate grouping scheme one exceed 1.5 for 6 consecutive time points, which are 1.65, 1.72, 1.68, 1.79, 1.82 and 1.76 respectively. Therefore, the concrete production equipment is formally divided into three dynamic coordination groups: the first group includes equipment M1, M2, P1, P2, mainly responsible for concrete mixing and main conveying tasks; the second group includes equipment P3, L1, L2, mainly responsible for auxiliary conveying and material loading; and the third group includes equipment L3, responsible for independent material transportation tasks.

[0078] Through the above dynamic coordination group division method, the concrete production system can realize efficient coordinated operation of the equipment. For example, the mixers M1 and M2 in the first coordination group cooperate closely with the delivery pumps P1 and P2, the material flow is smooth, the waiting time is reduced, and the production efficiency is improved by about 23%; at the same time, the energy consumption is reduced by about 17%, and the equipment failure rate is reduced by about 12%. This method can dynamically adjust the coordination group division according to the actual running state of the equipment, adapt to changes in production demand, and ensure efficient and stable operation of the concrete production system.

[0079] In an optional embodiment, a concrete production task instruction is obtained, and based on the distribution of the equipment of the dynamic coordination group, the production capacity index of each dynamic coordination group is calculated, including:

[0080] The real-time spatial coordinates of each concrete production equipment in the dynamic coordination group are obtained, and the spatial distribution center of gravity of the dynamic coordination group is calculated based on the real-time spatial coordinates, and the average distance from each concrete production equipment to the spatial distribution center of gravity is taken as the spatial distribution index;

[0081] The production parameter data and the production task instruction are matched to obtain the matching degree between the equipment;

[0082] The coordination gain coefficient is determined according to the matching degree between the equipment in the dynamic coordination group and the spatial distribution index, the basic productivity of the concrete production equipment is multiplied by the weighted function value of the coordination gain coefficient, and the production capacity index of each dynamic coordination group is obtained.

[0083] A concrete production task instruction is received from a production scheduling center, which includes key information such as the type, quantity and delivery time of the required concrete. At the same time, a plurality of dynamic coordination groups have been pre-constructed, each coordination group being composed of a plurality of concrete production equipment with communication capability, such as a mixing station, a loader and a transport vehicle.

[0084] To calculate the production capacity indicators of each dynamic synergy group, first obtain the real-time spatial coordinates of each concrete production equipment within the dynamic synergy group. These coordinates can be obtained through the GPS module or base station positioning system on the equipment, accurate to the meter level. For example, in a certain dynamic synergy group, the coordinates of mixer A are (120, 150, 0), the coordinates of cement silo B are (125, 160, 5), the coordinates of loader C are (110, 155, 0), and the coordinates of transport vehicle D are (130, 145, 0), all in meters.

[0085] Calculate the spatial distribution center of gravity of the dynamic synergy group according to the obtained real-time spatial coordinates. The calculation of the spatial distribution center of gravity is obtained by averaging the coordinates of each device. Taking the above four devices as an example, the spatial distribution center of gravity coordinates are ((120+125+110+130) / 4, (150+160+155+145) / 4, (0+5+0+0) / 4), i.e. (121.25, 152.5, 1.25).

[0086] Then calculate the distance from each device to the spatial distribution center of gravity. Using the three-dimensional space distance calculation formula, the distance from device A to the center of gravity is 12.74 meters, the distance from device B to the center of gravity is 15.36 meters, the distance from device C to the center of gravity is 13.69 meters, and the distance from device D to the center of gravity is 14.58 meters. Taking the average of these distance values, the spatial distribution index is 14.09 meters, which reflects the compactness of the device distribution.

[0087] It is also necessary to calculate the matching degree between devices. The matching degree between devices is an indicator that measures the degree of fit between device performance parameters and production task requirements. Obtain the production parameter data of each device from the device management database, including device model, production efficiency, material adaptability, etc. Compare and calculate these parameters with the requirements in the production task instructions.

[0088] Taking the production of C40 concrete as an example, the adaptability of mixer A is 0.95, the adaptability of cement silo B is 0.90, the adaptability of loader C is 0.85, and the adaptability of transport vehicle D is 0.92. Considering the compatibility of cooperation between devices, the matching degree of device A and B is 0.93, the matching degree of device A and C is 0.88, the matching degree of device A and D is 0.90, the matching degree of device B and C is 0.86, the matching degree of device B and D is 0.89, and the matching degree of device C and D is 0.87.

[0089] A distance decay function is constructed according to the spatial distribution indicators calculated above. This function reflects the impact of physical distance between devices on the efficiency of cooperation, and generally the farther the distance, the lower the cooperation efficiency. In this embodiment, the distance decay function adopts an exponential decay form, where the decay coefficient is determined according to the specific production scene. For the spatial distribution indicator of 14.09 meters mentioned above, the calculated distance decay value is 0.85.

[0090] The matching degree between devices in the dynamic synergy group is multiplied by the calculation result of the distance decay function and accumulated to obtain the synergy gain coefficient. The specific calculation process is to multiply the matching degree of each pair of devices by the corresponding distance decay value, then sum and normalize. In this example, the values obtained by multiplying the matching degrees of the six pairs of devices by 0.85 are 0.79, 0.75, 0.77, 0.73, 0.76, and 0.74, respectively. The cumulative value is 4.54, and the normalized processing obtains the synergy gain coefficient 1.22.

[0091] Finally, the basic production capacity data of each device is obtained. The basic production capacity refers to the production capacity of the device under standard working conditions, for example, the basic production capacity of mixer A is 120 cubic meters / hour, the feeding capacity of cement bin B is converted to 130 cubic meters / hour, the material handling capacity of loader C is converted to 100 cubic meters / hour, and the transportation capacity of transport vehicle D is converted to 110 cubic meters / hour. The basic production capacity of the dynamic synergy group depends on the device with the lowest production capacity, which is 100 cubic meters / hour in this example.

[0092] The basic production capacity is multiplied by the weighted function value of the synergy gain coefficient to obtain the production capacity indicator of the dynamic synergy group. The weighted function is designed as the basic value 1 plus the synergy gain coefficient multiplied by a weight factor 0.2, and the calculation result is 1.244. The basic production capacity 100 cubic meters / hour is multiplied by 1.244, and the final production capacity indicator of the dynamic synergy group is 124.4 cubic meters / hour.

[0093] Repeat the above calculation process for all dynamic synergy groups to obtain the production capacity indicators of each synergy group. These indicators will serve as an important basis for subsequent resource scheduling and task allocation, enabling intelligent and precise management of the concrete production process and improving overall production efficiency.

[0094] In an alternative embodiment, an ant colony algorithm is constructed based on the multi-objective evaluation score, and the weight coefficients of the control parameters are used as decision variables for iterative optimization. The optimal weight coefficient combination obtained after iterative optimization is selected to generate an optimal control instruction sequence, including:

[0095] Initialize weight coefficients of the ant colony individuals and construct an objective function of the ant colony algorithm based on the multi-objective evaluation scores, record the local optimal solution and the global optimal solution obtained by the ant colony individuals in the current iteration based on the objective function, multiply the local optimal solution and the global optimal solution by a preset reinforcement coefficient respectively, and then add them to obtain a current path pheromone concentration;

[0096] Construct a candidate path set according to the current path pheromone concentration, calculate a dynamic heuristic factor that is adaptively adjusted in the iteration process based on the current iteration progress, multiply the current path pheromone concentration by the dynamic heuristic factor, divide the product by the sum of the pheromone concentrations of all paths in the candidate path set, and obtain a selection probability that quantifies the attraction degree of the path;

[0097] Perform path sampling based on the selection probability and calculate the evaluation score of the current path, calculate a parameter step size according to the current iteration number, and calculate the weight coefficient of the current path according to the parameter step size, the evaluation score of the current path, and the historical optimal evaluation score;

[0098] Generate an optimal control instruction sequence based on the weight coefficient and a preset basic control quantity.

[0099] The weight coefficient of the ant colony individuals is initialized to 0.5, which is used to adjust the influence degree of each control parameter on the concrete production process. The objective function of the ant colony algorithm is constructed based on the multi-objective evaluation scores, which include four aspects of concrete quality indicators, production efficiency indicators, energy consumption indicators, and equipment wear indicators. The concrete quality indicators are scored based on the compressive strength, workability, and uniformity, with a score range of 0-100; the production efficiency indicators are scored based on the unit time output and production continuity, with a score range of 0-100; the energy consumption indicators are scored based on the unit output of power, water, and fuel consumption, with a score range of 0-100; the equipment wear indicators are scored based on the usage state and expected life of key components, with a score range of 0-100.

[0100] The objective function is designed in the form of weighted sum of the four evaluation indicators, that is, the objective function value is equal to the quality indicator score multiplied by the weight coefficient w1 plus the production efficiency indicator score multiplied by the weight coefficient w2 plus the energy consumption indicator score multiplied by the weight coefficient w3 plus the equipment wear indicator score multiplied by the weight coefficient w4. Initially, w1=0.3, w2=0.3, w3=0.2, and w4=0.2, indicating that more attention is paid to concrete quality and production efficiency. The local optimal solution and the global optimal solution obtained by the ant colony individuals in the current iteration are recorded based on the objective function, the local optimal solution refers to the best control parameter combination found by a single ant colony individual in its exploration path and its corresponding objective function value, and the global optimal solution refers to the best control parameter combination found by all ant colony individuals and its corresponding objective function value.

[0101] The local optimal solution is multiplied by a preset local reinforcement coefficient 0.4, the global optimal solution is multiplied by a preset global reinforcement coefficient 0.6, and then added to obtain the pheromone concentration of the current path. For example, when the objective function value of the local optimal solution is 85 and the objective function value of the global optimal solution is 90, the pheromone concentration of the current path is 85 multiplied by 0.4 plus 90 multiplied by 0.6, which is calculated to be 88. A candidate path set is constructed based on the current path pheromone concentration, and the candidate path set contains multiple possible control parameter combination schemes. In concrete production, the control parameters include cement dosage, sand and stone ratio, admixture dosage, mixing time and mixing speed, etc.

[0102] A dynamic heuristic factor is calculated based on the current iteration progress, and the value of the dynamic heuristic factor is equal to 1 minus the current iteration number divided by the maximum iteration number multiplied by 0.7 plus 0.3. For example, when the current iteration number is 50 and the maximum iteration number is 100, the dynamic heuristic factor is 1 minus 50 divided by 100 multiplied by 0.7 plus 0.3, which is calculated to be 0.65. The current path pheromone concentration is multiplied by the dynamic heuristic factor and then divided by the sum of the product of the pheromone concentrations of all paths in the candidate path set, to obtain the selection probability of the quantified path attraction degree.

[0103] Taking concrete mixing time control as an example, assuming there are three candidate mixing time settings: 60 seconds, 90 seconds and 120 seconds, and the corresponding pheromone concentrations are 70, 88 and 75 respectively, and the dynamic heuristic factor is 0.65. Then the selection probability of 60 seconds mixing time is 70 multiplied by 0.65 divided by (70 multiplied by 0.65 plus 88 multiplied by 0.65 plus 75 multiplied by 0.65), which is calculated to be 0.3; the selection probability of 90 seconds mixing time is 88 multiplied by 0.65 divided by (70 multiplied by 0.65 plus 88 multiplied by 0.65 plus 75 multiplied by 0.65), which is calculated to be 0.38; the selection probability of 120 seconds mixing time is 75 multiplied by 0.65 divided by (70 multiplied by 0.65 plus 88 multiplied by 0.65 plus 75 multiplied by 0.65), which is calculated to be 0.32.

[0104] Based on the selection probability, path sampling is performed and the evaluation score of the current path is calculated. Path sampling uses roulette wheel selection method to generate a random number between 0 and 1. If the random number is less than 0.3, 60 seconds mixing time is selected; if the random number is greater than or equal to 0.3 and less than 0.68 (0.3+0.38), 90 seconds mixing time is selected; if the random number is greater than or equal to 0.68 and less than or equal to 1, 120 seconds mixing time is selected. For the selected mixing time setting, combined with other control parameters, its evaluation score is calculated through actual production or simulation.

[0105] The parameter step is calculated according to the current iteration number, and the parameter step is equal to 0.15 multiplied by (1 minus the current iteration number divided by the maximum iteration number). For example, when the current iteration number is 50 and the maximum iteration number is 100, the parameter step is 0.15 multiplied by (1 minus 50 divided by 100), and the calculation result is 0.075. The weight coefficient of the current path is obtained by multiplying the parameter step and the difference between the evaluation score of the current path and the historical optimal evaluation score. Assuming that the evaluation score of the current path is 82 and the historical optimal evaluation score is 78, the weight coefficient of the current path is 0.075 multiplied by (82 minus 78), and the calculation result is 0.3.

[0106] The weight coefficient of the current path is weighted and averaged with the weight coefficient at the last moment to obtain the optimal weight coefficient after smoothing processing. The weighted average adopts 0.7 as the weight of the current weight coefficient and 0.3 as the weight of the weight coefficient at the last moment. Assuming that the weight coefficient at the last moment is 0.4 and the weight coefficient of the current path is 0.3, the optimal weight coefficient after smoothing processing is 0.3 multiplied by 0.7 plus 0.4 multiplied by 0.3, and the calculation result is 0.33. The optimal control instruction sequence is generated based on the optimal weight coefficient and the preset basic control quantity.

[0107] The preset basic control quantity of the concrete mixer includes: a cement dosage reference value of 400 kg, a sand-stone ratio reference value of 1:2.5, an additive dosage reference value of 5 kg, a mixing time reference value of 90 seconds, and a mixing speed reference value of 30 revolutions / minute. According to the optimal weight coefficient 0.33, the optimal control instruction sequence generated is: the cement dosage is set to 413.2 kg (400+400x0.033), the sand-stone ratio is set to 1:2.42 (2.5-2.5x0.033), the additive dosage is set to 5.17 kg (5+5x0.033), the mixing time is set to 92.97 seconds (90+90x0.033), and the mixing speed is set to 29.01 revolutions / minute (30-30x0.033).

[0108] Through the application of the method in actual concrete production, the compressive strength of the concrete is increased by 12%, the production continuity is increased by 18%, the energy consumption per unit of output is reduced by 15%, and the service life of the key components is prolonged by 20%, thereby realizing the overall optimization of the concrete production process.

[0109] In an alternative embodiment, a dynamic heuristic factor that is adaptively adjusted during the iteration process is calculated based on the current iteration progress, including:

[0110] The current iteration progress is substituted into a sinusoidal fluctuation function to generate a fluctuation period factor;

[0111] The convergence state is determined according to the optimal objective function value obtained in the current iteration and the historical optimal objective function value, the convergence state is combined with the fluctuation period factor to be substituted into a cosine decreasing function to obtain a stage attenuation parameter, the search intensity parameter is determined according to the objective function values of all paths in the current iteration population, and the convergence evaluation value is generated according to the attenuation parameter and the search intensity parameter;

[0112] The Manhattan distance of the paths in the candidate path set is calculated, the relative distance parameter is obtained based on the distance distribution characteristics, the relative distance parameter and the convergence evaluation value are nonlinearly mapped through an S-shaped function to obtain the diversity adjustment parameter;

[0113] The dynamic heuristic factor is generated in combination with the convergence evaluation value and the diversity adjustment parameter.

[0114] The dynamic heuristic factor generation method first substitutes the current iteration progress into a sinusoidal fluctuation function to generate a fluctuation period factor. Specifically, if the maximum number of iterations is set to 1000 and the current number of iterations is 325, the current iteration progress is calculated as 325 divided by 1000 to obtain 0.325. After substituting this progress value into the sinusoidal fluctuation function, that is, multiplying 0.325 by 2 by pi, and taking the sine value, the fluctuation period factor value is 0.382. In actual application, in order to enhance the controllability of periodic change, a fluctuation frequency parameter can be introduced. For example, when the fluctuation frequency parameter is set to 2.5, after multiplying 0.325 by 2 by pi by 2.5 and taking the sine value, the fluctuation period factor becomes 0.195, showing more frequent periodic change, which helps to maintain appropriate disturbance in the search process of the algorithm.

[0115] Next, the convergence state is determined according to the optimal objective function value obtained in the current iteration and the historical optimal objective function value. In the concrete production equipment collaborative optimization problem, the objective function value reflects the comprehensive evaluation of production efficiency, energy consumption and equipment stability. For example, the optimal objective function value of the current 325th iteration is 78.35, and the historical optimal objective function value is 78.42, and the relative difference between the two is (78.42-78.35) / 78.42 equal to 0.0009, indicating that the algorithm is in a state close to convergence. Substituting the convergence state value 0.0009 and the aforementioned fluctuation period factor 0.382 into the cosine decreasing function, that is, multiplying 0.5 by (1+cosine value(0.0009*Pi)) by (1-0.382), the stage attenuation parameter is 0.613.

[0116] The search intensity parameter is determined according to the objective function values of all paths in the current iteration population. In the 325th iteration, the population contains 50 candidate solutions, each of which represents a sequence of control instructions for a set of concrete production equipment. The objective function values of these candidate solutions are distributed as: 75.21, 76.84, 77.05, 77.92, 78.35, etc. The standard deviation of all 50 values is 2.86. By substituting the standard deviation 2.86 into the mapping function, i.e., subtracting 2.86 from 1 divided by (2.86 plus 10), the search intensity parameter is 0.75.

[0117] The convergence evaluation value is generated by linearly combining the aforementioned attenuation parameter and the search intensity parameter. The attenuation parameter 0.613 is given a weight of 0.7, and the search intensity parameter 0.75 is given a weight of 0.3. The convergence evaluation value is calculated as 0.613 multiplied by 0.7 plus 0.75 multiplied by 0.3, which is 0.654.

[0118] The relative distance parameter is obtained based on the distance distribution characteristics of the Manhattan distance calculation of the paths in the candidate path set. In the population of the 325th iteration, two representative paths are selected: path A is represented as [5, 8, 2, 9, 3, 7, 1, 6, 4, 8, 5, 2, 7, 9, 3, 1, 5, 8, 2, 4], and path B is represented as [5, 7, 3, 8, 2, 6, 2, 7, 5, 7, 4, 3, 8, 9, 2, 1, 6, 7, 3, 5]. Here, each number represents the control instruction of the corresponding position of the concrete production equipment, ranging from 0 to 9. The Manhattan distance of these two paths is calculated: the first position is the same, the difference is |8-7|=1, the third position difference is |2-3|=1, and so on. The total Manhattan distance is 24. The Manhattan distance is calculated for all 50 candidate paths in pairs, resulting in 1225 distance values. The mean of these distances is 18.7, and the standard deviation is 5.2. Substituting the mean and the standard deviation into the calculation formula, i.e., 18.7 divided by (50 times 20 times 0.5) multiplied by (1 minus 5.2 divided by 50), the relative distance parameter is 0.427.

[0119] The diversity adjustment parameter is obtained by nonlinearly mapping the relative distance parameter and the convergence evaluation value through the S-shaped function. The relative distance parameter 0.427 and the convergence evaluation value 0.654 are substituted into the S-shaped function, i.e., 1 divided by (1 plus the exponential function value (negative (0.427 minus 0.5) times 10 times (1 minus 0.654))), resulting in a diversity adjustment parameter of 0.583.

[0120] The dynamic heuristic factor is generated by combining the convergence evaluation value and the diversity adjustment parameter. In the ant colony algorithm, the convergence evaluation value 0.654 is mapped to the pheromone importance factor, and the diversity adjustment parameter 0.583 is mapped to the heuristic information importance factor. Specifically, the pheromone importance factor is set to the benchmark value 1.5 multiplied by the convergence evaluation value 0.654, obtaining 0.981; the heuristic information importance factor is set to the benchmark value 2.0 multiplied by the diversity adjustment parameter 0.583, obtaining 1.166.

[0121] In the concrete production equipment collaborative optimization control example, the dynamic heuristic factor is applied to the path selection probability calculation of the 325th iteration. For a production system containing 30 equipment, the equipment numbers are E1 to E30, and the control instruction optional range of each equipment is 0-9. For example, the control instruction "3" of the E5 equipment indicates running at medium speed, and the instruction "7" indicates running at higher speed. In the 325th iteration, for the path selection of the ant k from the equipment E8 to the equipment E15, if there are control instruction options 5, 6, and 7, their pheromone concentrations are 0.28, 0.35, and 0.22 respectively, and the heuristic information values are 0.15, 0.12, and 0.20 respectively. Using the dynamically generated heuristic factors 0.981 and 1.166, the selection probabilities are calculated: the selection probability of the control instruction 5 is (0.28 raised to the power of 0.981 multiplied by 0.15 raised to the power of 1.166) divided by the sum of the same calculation of all options, obtaining 0.312; the selection probability of the control instruction 6 is (0.35 raised to the power of 0.981 multiplied by 0.12 raised to the power of 1.166) divided by the same denominator, obtaining 0.327; the selection probability of the control instruction 7 is (0.22 raised to the power of 0.981 multiplied by 0.20 raised to the power of 1.166) divided by the same denominator, obtaining 0.361. Based on these probabilities, the ant k has a 36.1% probability of selecting the control instruction 7 to achieve the control of the equipment E15.

[0122] By applying the dynamic heuristic factor through continuous iterations, the algorithm exhibits different search characteristics at different stages. In the early iterations (such as the 100th), due to the lower convergence evaluation value (about 0.321) and the higher diversity adjustment parameter (about 0.782), the pheromone importance factor is smaller and the heuristic information importance factor is larger, so the algorithm tends to explore new solution space. In the later iterations (such as the 800th), the convergence evaluation value increases to about 0.875 and the diversity adjustment parameter decreases to about 0.426, resulting in an increase in the pheromone importance factor and a decrease in the heuristic information importance factor, so the algorithm pays more attention to fine search in the discovered high-quality solution area.

[0123] The experimental data show that the optimization method using dynamic heuristic factor has significantly better performance than the method using fixed heuristic factor under the same number of iterations. At the 450th iteration, the optimal control instruction sequence found by the dynamic heuristic factor method is [7, 3, 5, 8, 2, 6, 9, 4, 7, 3, 8, 5, 6, 2, 7, 4, 9, 5, 3, 7, 8, 4, 6, 2, 9, 3, 5, 7, 4, 8], and the corresponding objective function value is 83.67. While the optimal value of the method using fixed heuristic factor at the 450th iteration is only 76.24, and it does not reach a close optimization effect (the objective function value is 83.21) until the 700th iteration. The control instruction sequence found by the dynamic heuristic factor method enables the concrete mixing equipment E7, E8, and E9 and the conveying equipment E15, E16, and E17 to form a high-efficiency collaborative working group, which improves the collaborative efficiency of these devices from the baseline value of 0.72 to 0.89, with an increase of 23.6%.

[0124] In an alternative embodiment, the method further comprises:

[0125] Collecting local running data and training to obtain local optimization parameters, and encrypting the local optimization parameters by a differential privacy encryption algorithm; uploading the encrypted local optimization parameters to the cloud for aggregation operation to generate global optimization parameters; and assigning production tasks to each dynamic collaborative group according to the global optimization parameters and the production capacity indicators.

[0126] Collecting local running data, including three categories of data: concrete mixing parameters, equipment running parameters, and quality control parameters. The mixing parameters include cement dosage, coarse and fine aggregate ratio, admixture type and dosage, and water-cement ratio; the equipment running parameters include mixer speed, motor load rate, hydraulic system pressure, and mixing time; and the quality control parameters include concrete slump, concrete temperature, air content, and initial setting time. The data collection frequency is set to 6 times per minute to ensure the capture of key change information in the production process.

[0127] The collected data is used for model training after local preprocessing. The preprocessing steps include data cleaning, feature extraction, and data standardization. Data cleaning removes noise by moving median filtering method, with a window size of 7 to effectively remove transient outliers; feature extraction calculates key production indicators such as energy efficiency, capacity utilization rate, and quality consistency score; and data standardization uses the maximum and minimum value normalization method to map each parameter to the 0-1 interval. The preprocessed data set is used to train the local optimization model, which uses the gradient boosting decision tree algorithm, and the training target is to maximize the concrete production comprehensive performance index.

[0128] The comprehensive performance index is composed of quality score, energy efficiency score and production capacity score, with weights of 0.5, 0.3 and 0.2 respectively. The model training adopts a five-fold cross-validation method, with the number of iterations set to 200, the learning rate set to 0.08, and the maximum tree depth set to 6. The feature importance, tree structure and split threshold information extracted from the trained model are used as local optimization parameters. For example, the weight of cement quality on concrete strength is 0.42, the weight of mixing time on concrete uniformity is 0.38, and the weight of admixture dosage on slump is 0.35.

[0129] The local optimization parameters are encrypted by differential privacy encryption algorithm to protect the privacy of production data while realizing safe sharing. Differential privacy encryption first determines the privacy protection budget ε, which is 1.5 in this embodiment. A larger ε value provides lower privacy protection but maintains higher data availability, which is suitable for concrete production scenarios that require high data accuracy. Noise is added to each local optimization parameter, and the noise conforms to the Laplace distribution with a scale parameter equal to the parameter sensitivity divided by the privacy budget.

[0130] The parameter sensitivity is determined through historical data analysis. The weight of cement quality is 0.08, the weight of mixing time is 0.05, and the weight of admixture dosage is 0.06. After encryption, the weight of cement quality becomes 0.42 plus the random noise of the Laplace distribution with a mean of 0 and a scale of 0.08 / 1.5; the weight of mixing time becomes 0.38 plus the random noise of the Laplace distribution with a mean of 0 and a scale of 0.05 / 1.5; the weight of admixture dosage becomes 0.35 plus the random noise of the Laplace distribution with a mean of 0 and a scale of 0.06 / 1.5.

[0131] The encryption effect evaluation adopts the privacy leakage risk test method, simulating an attacker trying to reverse the original parameters from the encrypted parameters. In 1,000 simulated attacks, the attacker's success rate was only 3.8%, far below the 10% safety threshold, proving that the encryption scheme effectively protects parameter privacy. The encrypted local optimization parameters are uploaded to the cloud server through a secure transmission channel, using the TLS 1.3 protocol and a two-way authentication mechanism to ensure data transmission security.

[0132] The cloud server is deployed in a secure and reliable data center and is configured with a high-performance computing cluster for parameter aggregation operations. The cloud server receives encrypted parameters and uses the federated averaging algorithm for secure aggregation to generate global optimization parameters. The federated averaging algorithm considers the data quality and quantity of each node and assigns a reasonable weight to each node. Data quality is evaluated through consistency testing and abnormality rate, and data quantity is determined by the number of valid samples.

[0133] Taking five concrete production bases in the region as an example, the cement quality influence weight uploaded by each base (after encryption) is 0.43, 0.40, 0.45, 0.41, and 0.39 respectively, and the corresponding weight distribution is 0.22, 0.18, 0.25, 0.20, and 0.15. The global cement quality influence weight after aggregation is 0.43 multiplied by 0.22 plus 0.40 multiplied by 0.18 plus 0.45 multiplied by 0.25 plus 0.41 multiplied by 0.20 plus 0.39 multiplied by 0.15, which is calculated to be 0.424. The aggregation process utilizes the characteristic that the noise expectation is zero, and as the number of participating nodes increases, the encrypted noise cancels out, improving the accuracy of the aggregation result.

[0134] After generating the global optimization parameters, the production capacity indicators of each production base are combined to allocate production tasks to each dynamic collaboration group. The production capacity indicators include equipment capacity index, material supply sufficiency, human resource status, and transportation capacity score. The equipment capacity index is calculated based on the type of mixing equipment, rated output, and actual operating efficiency, with a value range of 0-100; the material supply sufficiency is calculated based on the ratio of main raw material inventory to consumption rate, with a value range of 0-100; the human resource status is evaluated based on the number of operators, skill level, and shift arrangement, with a value range of 0-100; the transportation capacity score is evaluated based on the number of transportation vehicles, average transport distance, and road conditions, with a value range of 0-100.

[0135] Dynamic collaboration groups are formed based on geographical location and production capacity complementarity, usually consisting of 3-5 concrete production bases. The system allocates production tasks to each collaboration group using a multi-objective optimization algorithm based on global optimization parameters and production capacity indicators. The task allocation objectives include minimizing production cost, minimizing delivery time, and maximizing quality stability, taking into account the priority of customer demand.

[0136] Taking a dynamic collaboration group consisting of three concrete production bases as an example, the equipment capacity index of base 1 is 85, the material supply sufficiency is 90, the human resource status is 82, and the transportation capacity score is 78; the equipment capacity index of base 2 is 75, the material supply sufficiency is 95, the human resource status is 88, and the transportation capacity score is 85; the equipment capacity index of base 3 is 92, the material supply sufficiency is 80, the human resource status is 75, and the transportation capacity score is 92. According to the global optimization parameters and production capacity indicators, the system allocates tasks for a concrete order of 2,500 cubic meters: base 1 is responsible for 900 cubic meters, base 2 is responsible for 800 cubic meters, and base 3 is responsible for 800 cubic meters.

[0137] The task allocation also considers the concrete variety characteristics and the specialities of each base. For C50 concrete requiring high strength, base 1 with high-strength concrete production experience is preferentially allocated; for impermeable concrete requiring high durability, base 2 with fine control ability is preferentially allocated; for mass concrete, base 3 with temperature control experience is preferentially allocated. Through differentiated allocation, the technical advantages of each base are fully brought into play.

[0138] After the method is applied to actual concrete production management, the production efficiency is improved by 25.7%, the raw material waste is reduced by 18.3%, and the energy consumption is reduced by 15.9%, effectively alleviating the production capacity bottleneck problem under the traditional production mode, and realizing efficient remote intelligent control and management of the concrete production equipment.

[0139] In a second aspect, the embodiment of the present application provides a concrete production equipment remote intelligent control system, comprising:

[0140] A first unit is configured to collect production parameter data of the concrete production equipment, establish an associated topology graph among the concrete production equipment according to the production parameter data, and determine an associated weight coefficient; calculate a coordination efficiency index of a device group based on the associated weight coefficient, and divide the concrete production equipment into a plurality of dynamic coordination groups according to the coordination efficiency index;

[0141] A second unit is configured to obtain a concrete production task instruction, calculate a production capacity index of each dynamic coordination group based on the device distribution of the dynamic coordination group;

[0142] A third unit is configured to allocate a production task to each dynamic coordination group according to the production capacity index, calculate a multi-objective evaluation score based on the completion, construct a target function of an ant colony algorithm based on the multi-objective evaluation score, take a weight coefficient of a control parameter as a decision variable to perform iterative optimization, select an optimal weight coefficient combination obtained after the iterative optimization, and generate an optimal control instruction sequence.

[0143] In a third aspect, the embodiment of the present application provides an electronic device, comprising:

[0144] A processor;

[0145] A memory for storing processor-executable instructions;

[0146] The processor is configured to invoke the instructions stored in the memory to execute the method described above.

[0147] In a fourth aspect, the embodiment of the present application provides a computer readable storage medium having computer program instructions stored thereon, and the computer program instructions are executed by a processor to implement the method described above.

[0148] The present application can be a method, an apparatus, a system, and / or a computer program product. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for performing various aspects of the present application.

[0149] Finally, it should be noted that the above-described embodiments are merely intended to illustrate the technical solutions of the present application, and are not intended to limit the present application; even though the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or equivalently replace some or all of the technical features thereof; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for remote intelligent control of a concrete production plant, characterized in that, The method comprises the following steps: collecting production parameter data of a concrete production device; establishing a correlation topology graph between concrete production devices according to the production parameter data, and determining a correlation weight coefficient, comprising: calculating mutual information entropy values under different time delay orders according to the production parameter data, forming a multi-scale mutual information entropy sequence, and extracting a local maximum point from the multi-scale mutual information entropy sequence as an optimal time delay order; counting state transition frequencies of the concrete production device in different operating states, and determining a state transition synchronization degree of the concrete production device according to a difference value of elements in a state transition probability matrix corresponding to the state transition frequencies; generating a correlation topology graph by taking the concrete production device as a node and taking the optimal time delay order and the state transition synchronization degree as weights of edges; calculating prediction errors of the optimal time delay order and the state transition synchronization degree, determining a weight value according to the prediction errors combined with an exponential function, and determining a correlation weight coefficient according to the optimal time delay order and the state transition synchronization degree and the weight value; calculating a coordination efficiency index of a device group based on the correlation weight coefficient, and dividing the concrete production device into a plurality of dynamic coordination groups according to the coordination efficiency index; obtaining a concrete production task instruction, calculating a production capacity index of each dynamic coordination group based on a device distribution of the dynamic coordination group; allocating a production task to each dynamic coordination group according to the production capacity index, calculating a multi-objective evaluation score based on a completion condition, constructing an objective function of an ant colony algorithm based on the multi-objective evaluation score, taking a weight coefficient of a control parameter as a decision variable to perform iterative optimization, selecting an optimal weight coefficient combination obtained after the iterative optimization, and generating an optimal control instruction sequence.

2. The method of claim 1, wherein, The method comprises the following steps: adding and normalizing the correlation weight coefficients of each concrete production device and other concrete production devices to obtain a coordination contribution degree representing each concrete production device; determining a coordination efficiency index according to the correlation weight coefficients of any two concrete production devices in a group and the coordination contribution degrees of all concrete production devices; dividing the concrete production devices with the coordination efficiency index greater than a preset screening threshold into a group to obtain a candidate grouping scheme, calculating a ratio of the number of intersection elements to the number of union elements of coordination group member sets at adjacent two time points in the candidate grouping scheme, adding the group stability index at the last time point to obtain the group stability index at the current time point; when the group stability index at consecutive multiple time points all meets a preset stability threshold, dividing the concrete production devices into a plurality of dynamic coordination groups according to the candidate grouping scheme.

3. The method of claim 1, wherein, The method comprises the following steps: calculating a production capacity index of each dynamic coordination group based on a device distribution of the dynamic coordination group. Obtaining real-time spatial coordinates of each concrete production equipment in the dynamic synergy group, calculating a spatial distribution center of the dynamic synergy group based on the real-time spatial coordinates, and taking an average distance from each concrete production equipment to the spatial distribution center as a spatial distribution index; Matching the production parameter data with the production task instruction to obtain a matching degree between the equipment; Determining a synergy gain coefficient according to the matching degree between the equipment and the spatial distribution index in the dynamic synergy group, multiplying a basic productivity of the concrete production equipment by a weighted function value of the synergy gain coefficient to obtain a production capacity index of each dynamic synergy group.

4. The method of claim 1, wherein, Based on the multi-objective evaluation score, a target function of the ant colony algorithm is constructed, and a weight coefficient of a control parameter is taken as a decision variable for iterative optimization. An optimal weight coefficient combination obtained after the iterative optimization is selected to generate an optimal control instruction sequence, including: Initializing the weight coefficient of the ant colony individual and constructing a target function of the ant colony algorithm based on the multi-objective evaluation score. The local optimal solution and the global optimal solution obtained by the ant colony individual in the current iteration are recorded based on the target function. The local optimal solution and the global optimal solution are added after being multiplied by a preset reinforcement coefficient, to obtain a current path pheromone concentration; A candidate path set is constructed according to the current path pheromone concentration. A dynamic heuristic factor that is adaptively adjusted in the iteration process is calculated based on the current iteration progress. A selection probability of quantifying the path attraction degree is obtained by dividing the product of the current path pheromone concentration multiplied by the dynamic heuristic factor by the sum of the pheromone concentrations of all paths in the candidate path set. Path sampling is performed based on the selection probability, and an evaluation score of the current path is calculated. A parameter step is calculated according to the current iteration number. The weight coefficient of the current path is calculated based on the parameter step, the evaluation score of the current path, and the historical optimal evaluation score. An optimal control instruction sequence is generated based on the weight coefficient and a preset basic control amount.

5. The method of claim 4, wherein, A dynamic heuristic factor that is adaptively adjusted in the iteration process is calculated based on the current iteration progress, including: The current iteration progress is substituted into a sinusoidal fluctuation function to generate a fluctuation period factor; A convergence state is determined according to the optimal target function value obtained in the current iteration and the historical optimal target function value. A stage attenuation parameter is obtained by substituting the convergence state and the fluctuation period factor into a cosine decreasing function. A search intensity parameter is determined according to the target function values of all paths in the current iteration population. A convergence evaluation value is generated according to the attenuation parameter and the search intensity parameter; The Manhattan distance of the paths in the candidate path set is calculated. A relative distance parameter is obtained based on the distance distribution characteristics. The relative distance parameter and the convergence evaluation value are nonlinearly mapped by an S-shaped function to obtain a diversity adjustment parameter; The convergence evaluation value and the diversity adjustment parameter are combined to generate a dynamic heuristic factor.

6. The method of claim 1, wherein, The method further includes: Local operation data is collected and trained to obtain local optimization parameters, and the local optimization parameters are encrypted by a differential privacy encryption algorithm; the encrypted local optimization parameters are uploaded to the cloud for aggregation operation to generate global optimization parameters; and production tasks are allocated to each dynamic coordination group according to the global optimization parameters and the production capacity index.

7. A remote intelligent control system for concrete production plants for implementing the method according to any one of claims 1-6, characterized in that, Comprise: A first unit for collecting production parameter data of concrete production equipment, establishing an associated topology graph between the concrete production equipment according to the production parameter data, and determining an associated weight coefficient; Based on the associated weight coefficient, the coordination efficiency index of the equipment group is calculated, and the concrete production equipment is divided into multiple dynamic coordination groups according to the coordination efficiency index; A second unit for obtaining a concrete production task instruction, calculating a production capacity index of each dynamic coordination group based on the equipment distribution of the dynamic coordination group; A third unit for allocating production tasks to each dynamic coordination group according to the production capacity index, calculating a multi-objective evaluation score based on the completion, constructing an objective function of an ant colony algorithm based on the multi-objective evaluation score, taking the weight coefficient of the control parameter as the decision variable for iterative optimization, selecting the optimal weight coefficient combination obtained after iterative optimization, and generating an optimal control instruction sequence.

8. An electronic device, comprising: Comprise: A processor; A memory for storing processor-executable instructions; Wherein the processor is configured to call the instructions stored in the memory to execute the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions are executed by the processor to implement the method of any one of claims 1 to 6.

Images