Equipment prediction maintenance and flexible production manufacturing equipment collaborative operation method

Abstract

The invention relates to the technical field of industrial production, in particular to an equipment prediction maintenance and flexible production manufacturing equipment collaborative operation method. According to the technical scheme, the method comprises the following steps of 1, collecting key parameters of an equipment state in real time, transmitting normalized data to a cloud platform for processing, 2, evaluating the health state of the equipment by using a health index according to the collected equipment data, and 3, determining the health state of the equipment by using historical equipment state data and fault records. The method comprises the steps of 1, establishing a fault prediction model by adopting an LSTM network machine learning algorithm, and predicting the time when equipment fails, 4, dynamically adjusting a production plan based on a prediction maintenance result, 5, issuing the optimized plan to a workshop production control system, and 6, evaluating the effectiveness of the collaborative operation method through indexes. According to the method, a prediction maintenance mechanism is introduced, potential fault risks of the equipment are found in advance, and through a flexible production manufacturing mode, stable operation of the equipment is ensured, and meanwhile, dynamic adjustment of production tasks is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial production, and particularly relates to a collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment. Background Art

[0002] In modern manufacturing systems, equipment maintenance strategies are mainly divided into planned maintenance and corrective maintenance. However, the traditional fixed-cycle maintenance method has certain limitations, which are prone to over-maintenance or under-maintenance, resulting in increased maintenance costs or unplanned equipment downtime. In addition, unexpected equipment failures not only affect the execution of production plans, but also may cause material waste and resource losses, reducing the overall competitiveness of enterprises.

[0003] The manufacturing industry has put forward higher requirements for the stability, reliability and availability of equipment. How to reduce maintenance costs while ensuring the efficient operation of equipment has become a difficult problem that manufacturing enterprises urgently need to solve.

[0004] Predictive maintenance analyzes the operation data of equipment, establishes degradation models and health assessment systems, and realizes dynamic monitoring and early warning of equipment status. This method uses machine learning, data mining and Internet of Things technologies to maintain equipment before a failure occurs, thus avoiding sudden downtime.

[0005] Flexible production manufacturing, as an important part of modern manufacturing models, emphasizes dynamically adjusting production tasks and equipment configurations so that the production system can quickly respond to market demands and changes in equipment status. Combining flexible production with predictive maintenance can effectively reduce the adverse effects brought by equipment failures on the premise of ensuring production continuity.

[0006] At present, some manufacturing enterprises have introduced predictive maintenance technologies, but mostly for single equipment or production lines, and the collaborative maintenance and production scheduling of equipment groups have not been realized. Facing a complex manufacturing environment, it is difficult to meet the needs of large-scale production only relying on the maintenance prediction of a single piece of equipment. Therefore, proposing a collaborative operation method that can combine equipment predictive maintenance and flexible production manufacturing is of great significance for enhancing the competitiveness of manufacturing enterprises.

[0007] Therefore, we propose a collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment to solve the existing problems. Summary of the Invention

[0008] The object of the present invention is to propose a collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment in view of the problems existing in the background art.

[0009] To achieve the above object, the present invention provides the following technical solution: A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment, including the following steps. Step 1: Key parameters of the equipment status are collected in real time, and the normalized data is transmitted to the cloud platform through a wireless network for processing. Step 2: According to the collected equipment data, the health status of the equipment is evaluated using a health index. Step 3: Using historical equipment status data and fault records, a fault prediction model is established using the LSTM network machine learning algorithm to predict the time when the equipment fails. Step 4: Based on the predictive maintenance results, the production plan is dynamically adjusted. Step 5: The optimized plan is sent to the workshop production control system. Step 6: The effectiveness of the collaborative operation method is evaluated through indicators.

[0010] Preferably, Step 1, namely equipment status data collection and processing, installs multiple types of sensors at key parts of key equipment, including temperature sensors, vibration sensors, pressure sensors, current sensors, etc. The sensor collection frequency is set to 10 times per second to ensure the real-time and continuity of data. In order to eliminate the dimensional differences of different sensors, the collected data needs to be normalized. Finally, the normalized data is transmitted to the central control system or cloud platform through an industrial wireless network or Ethernet and stored in the database according to the time stamp.

[0011] Preferably, Step 2, namely equipment health status evaluation, includes feature extraction, health index calculation, and status classification. Feature extraction refers to extracting equipment status features from the collected data, such as frequency domain features (root mean square value, peak value) of vibration signals, temperature fluctuation range, current stability, etc. Health index calculation refers to calculating the health index based on equipment features. Status classification refers to classifying the equipment status into three levels according to the health index.

[0012] Preferably, Step 3, namely fault prediction model establishment, uses historical data and fault records to fit the equipment degradation model to describe the degradation process of equipment performance over time. When the degradation degree reaches the threshold, the equipment is predicted to fail, and the fault time is calculated. By analyzing the degradation model, the time node of equipment failure is calculated in advance to facilitate arranging maintenance work in advance and reducing the risk of unplanned downtime.

[0013] Preferably, Step 4, namely dynamic adjustment of the flexible production plan, includes setting constraint conditions and setting optimization goals. Setting constraint conditions incorporates the equipment health status and fault prediction results into the production plan constraints to ensure that high-risk equipment is given priority to arrange maintenance tasks, and low-risk equipment undertakes the main production tasks. Setting optimization goals refers to allocating tasks through an optimization algorithm, and the objective function is to maximize production efficiency.

[0014] Preferably, in step 5, i.e., collaborative operation execution, after the optimized plan is sent to the workshop production control system, the equipment execution follows the following process: Equipment in good health status gives priority to executing high-priority tasks to maximize production capacity and resource utilization rate and ensure the timely completion of key production tasks; Equipment in a warning state undertakes short-cycle or low-risk tasks, avoids long-term high-load operation, and arranges for early maintenance to prevent the state from deteriorating further. The maintenance work and production tasks are carried out alternately to minimize downtime; Equipment in poor health status stops running immediately, and maintenance and repair measures are implemented preferentially. After the maintenance is completed and the equipment returns to a healthy state, the equipment can be put back into production. During this process, the production plan is adjusted dynamically, and other equipment shares its tasks to ensure that the overall production progress is not significantly affected.

[0015] Preferably, in step 6, i.e., maintenance and production result evaluation, the equipment maintenance and production result evaluation includes three core indicators: equipment utilization rate, average maintenance time, and unplanned downtime rate. The equipment utilization rate measures the ratio of the actual running time of the equipment to the total time, reflecting the operating efficiency of the equipment. The average maintenance time calculates the average value of each maintenance time to help evaluate the timeliness and efficiency of the maintenance work. The unplanned downtime rate calculates the proportion of the unplanned downtime time in the total time and is an important indicator for evaluating the system stability. Comprehensive analysis of these indicators helps to timely discover the weak links in the equipment operation, optimize the maintenance strategy, and improve the equipment reliability and production efficiency.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0017] Reduce the unplanned downtime rate: Through equipment status monitoring and fault prediction, equipment maintenance is carried out in advance to avoid sudden equipment failures, effectively reduce unplanned downtime time, and improve the operation stability of the equipment;

[0018] Improve the equipment utilization rate: Dynamically allocate production tasks, so that equipment in good health status undertakes the main production tasks, while equipment in a warning or poor state is arranged for maintenance or to execute low-load tasks, maximizing the overall utilization rate of the equipment;

[0019] Optimize the allocation of maintenance resources: Flexibly arrange the maintenance time and intensity according to different equipment states, avoid unnecessary regular maintenance or over-maintenance, thereby saving maintenance costs and improving the pertinence and efficiency of the maintenance work;

[0020] Enhance production flexibility: The production tasks are adjusted in real time according to the equipment status. Even if a certain piece of equipment stops running due to a fault, other equipment can timely share the production tasks to ensure the smooth progress of the production plan and reduce the impact of downtime on the overall production.

[0021] Enhance system intelligence: By introducing machine learning and self-learning algorithms, the equipment fault prediction model is continuously optimized during operation, improving the accuracy and real-time performance of fault prediction, enabling the manufacturing system to have the ability of continuous learning and self-optimization. Brief Description of the Drawings

[0022] Figure 1 is a schematic structural diagram of the present invention; Detailed Embodiments

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0024] Embodiment 1

[0025] As Figure 1 , a collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment proposed by the present invention includes the following steps. Step 1: Key parameters of the equipment status are collected in real time, and the normalized data is transmitted to the cloud platform through a wireless network for processing. Step 2: According to the collected equipment data, the health status of the equipment is evaluated using a health index. Step 3: Using historical equipment status data and fault records, a fault prediction model is established using the LSTM network machine learning algorithm to predict the time when the equipment fails. Step 4: Based on the predictive maintenance results, the production plan is dynamically adjusted. Step 5: The optimized plan is sent to the workshop production control system. Step 6: The effectiveness of the collaborative operation method is evaluated through indicators.

[0026] The collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment proposed by the present invention aims to solve problems such as unplanned downtime, over-maintenance or under-maintenance, and low production efficiency caused by sudden equipment failures in traditional manufacturing systems. By introducing a predictive maintenance mechanism, potential equipment failure risks can be detected in advance and maintenance can be carried out at the appropriate time, which not only reduces downtime but also effectively extends the equipment life. In addition, through the flexible production manufacturing mode, while ensuring the stable operation of the equipment, the production tasks can be dynamically adjusted, improving the flexibility and response ability of the production system.

[0027] This embodiment combines advanced technologies such as the Internet of Things, machine learning, and big data analysis. Multiple types of sensors are used to monitor the key components of the equipment in real time. The collected status parameter data is processed and analyzed. By establishing an equipment health assessment and fault prediction model, accurate equipment status information is provided for the production system. Based on the predictive maintenance results, the production plan is dynamically adjusted, and finally the optimal allocation of maintenance resources and production tasks is achieved.

[0028] Step 1: Acquisition and Processing of Equipment Status Data

[0029] The real-time acquisition of equipment status data is the basic work for implementing predictive maintenance. To ensure the comprehensiveness and accuracy of monitoring data, multiple types of sensors are installed at key parts of the equipment, including temperature sensors, vibration sensors, pressure sensors, current sensors, etc. These sensors can cover the core parameters of equipment operation and comprehensively reflect the equipment operation status;

[0030] The acquisition frequency of the sensors is set to 10 times per second, and the data accuracy reaches three decimal places to effectively capture the subtle changes in the equipment operation status. For example, on a high-speed rotating motor or transmission device, tiny vibrations or temperature changes may indicate that internal components of the equipment are about to wear or be damaged. Therefore, high-frequency data acquisition is crucial to ensure the real-time and continuity of data. The data dimensions and ranges collected by different sensors are different. For example, vibration signals are represented by acceleration (m / s 2 ), temperature is represented by degrees Celsius, and current is represented by amperes. To ensure the comparability of data with different dimensions in the same analysis model, the collected data needs to be normalized. The collected data is normalized through the following formula:

[0031]

[0032] Where:

[0033] x is the original data;

[0034] x min and x max are the minimum and maximum values of the data respectively;

[0035] x ′ is the normalized data.

[0036] Finally, the normalized data is transmitted to the central control system or cloud platform through an industrial wireless network or Ethernet and stored in the database according to the timestamp.

[0037] Step 2 Equipment Health Status Assessment

[0038] It includes feature extraction, health index calculation, and status classification. Feature extraction refers to extracting equipment status features from the collected data, such as the frequency domain features (root mean square value, peak value) of vibration signals, temperature fluctuation range, current stability, etc. Health index calculation refers to calculating the health index based on equipment features. The health index calculation formula is:

[0039] HI = w1·f1(x ′ ) + w2·f2(x ′ ) + … + w n·f n (x ′ )

[0040] where:

[0041] f1(x ′ ) is the characteristic index function of the equipment status;

[0042] w n is the weight of the characteristic index, satisfying

[0043] The larger the HI value, the better the equipment status.

[0044] Status classification means classifying the equipment status into three levels according to the health index:

[0045] Healthy (HI > 0.8);

[0046] Warning (0.5 ≤ HI ≤ 0.8);

[0047] Fault (HI ≤ 0.5).

[0048] Step 3: Establishment of the fault prediction model

[0049] Use historical data and fault records to fit the equipment degradation model. The degradation function of the equipment is:

[0050] D(t) = D0 + αt + βt 2

[0051] where:

[0052] D0 is the initial state;

[0053] α and β are the equipment degradation rate parameters;

[0054] t is the time.

[0055] Describe the degradation process of the equipment performance over time. When the degradation degree reaches the threshold, the equipment is predicted to fail, and the fault time is calculated:

[0056] Predict the fault time T of the equipment by analyzing the relationship between D(t) and the fault threshold D threshold : f

[0057] T f = min{t | D(t) ≥ D threshold}

[0058] By analyzing the degradation model, calculate the time node of equipment failure in advance, which is convenient for arranging maintenance work in advance and reducing the risk of unplanned downtime.

[0059] Step 4: Dynamic adjustment of the flexible production plan ​

[0060] The dynamic adjustment of the flexible production plan includes setting constraint conditions and optimization objectives. The setting of constraint conditions incorporates the equipment health status and fault prediction results into the production plan constraints, ensuring that high-risk equipment is given priority in arranging maintenance tasks, and low-risk equipment undertakes the main production tasks. The setting of optimization objectives refers to allocating tasks through an optimization algorithm, with the objective function being to maximize production efficiency. The mathematical expression of the optimization algorithm is as follows:

[0061]

[0062] Where:

[0063] c ij is the priority of task i executed on equipment j;

[0064] x ij is the decision variable (0 or 1) for allocating task i to equipment j.

[0065] Step 5: Collaborative operation execution

[0066] After the optimized plan is issued to the workshop production control system, the equipment executes according to the following process:

[0067] Equipment with good health status gives priority to executing high-priority tasks to maximize production capacity and resource utilization rate, ensuring the timely completion of key production tasks;

[0068] Equipment in a warning state undertakes short-cycle or low-risk tasks, avoiding long-term high-load operation, and arranging early maintenance to prevent the state from deteriorating further. The maintenance work and production tasks are carried out alternately to minimize the downtime;

[0069] Equipment with poor health status immediately stops running, and priority is given to implementing repair and maintenance measures. After the maintenance is completed and the health status is restored, the equipment can be put back into production;

[0070] During this process, the production plan is dynamically adjusted, and other equipment shares its tasks to ensure that the overall production schedule is not significantly affected.

[0071] Step 6: Maintenance and production result evaluation

[0072] The evaluation of equipment maintenance and production results includes three core indicators: equipment utilization rate, average maintenance time, and unplanned downtime rate. The equipment utilization rate measures the ratio of the actual running time of the equipment to the total time, reflecting the operating efficiency of the equipment. The mathematical expression of the equipment utilization rate is as follows:

[0073]

[0074] The average maintenance time calculates the average value of each maintenance time, helping to evaluate the timeliness and efficiency of the maintenance work. The mathematical expression of the average maintenance time is as follows:

[0075]

[0076] The unplanned shutdown rate calculates the proportion of unplanned shutdown time in the total time. The mathematical expression of the unplanned shutdown rate is as follows:

[0077]

[0078] As an important indicator for evaluating the system stability, comprehensively analyzing these indicators helps to promptly identify the weak links in the equipment operation, optimize the maintenance strategy, and improve the equipment reliability and production efficiency.

[0079] The above specific embodiments are only several preferred embodiments of the present invention. Based on the technical solution of the present invention and the relevant inspirations of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

[0080] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes falling within the meaning and scope of the equivalent elements of the claims in the present invention.

Claims

1. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment, comprising the following steps, characterized in that: Step 1: Collect key parameters of the device status in real time. The normalized data is transmitted to the cloud platform via a wireless network for processing. Step 2: Based on the collected device data, evaluate the health status of the device using a health index. Step 3: Use historical device status data and fault records to establish a fault prediction model with the LSTM network machine learning algorithm to predict the time of device failure. Step 4: Dynamically adjust the production plan based on the predictive maintenance results. Step 5: Send the optimized plan to the workshop production control system. Step 6: Evaluate the effectiveness of the collaborative operation method through indicators.

2. The collaborative operation method of equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: Step 1, namely device status data acquisition and processing, installs multiple types of sensors at key parts of key devices, including temperature sensors, vibration sensors, pressure sensors, current sensors, etc. The sensor acquisition frequency is set to 10 times per second to ensure the real-time and continuity of data. To eliminate the dimensional differences of different sensors, the collected data needs to be normalized. Finally, the normalized data is transmitted to the central control system or cloud platform via an industrial wireless network or Ethernet and stored in the database according to the timestamp.

3. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: Step 2, namely device health status evaluation, includes feature extraction , health index calculation, and status classification. Feature extraction refers to extracting device status features from the collected data, such as frequency domain features (root mean square value, peak value) of vibration signals, temperature fluctuation range, current stability, etc. Health index calculation refers to calculating the health index based on device features. Status classification refers to classifying the device status into three levels according to the health index.

4. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: Step 3, namely fault prediction model establishment, uses historical data and fault records to fit the device degradation model to describe the degradation process of device performance over time. When the degradation degree reaches the threshold, the device is predicted to fail, and the fault time is calculated. By analyzing the degradation model, the time node of device failure is calculated in advance to facilitate arranging maintenance work in advance and reducing the risk of unplanned downtime.

5. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: Step 4, namely dynamic adjustment of the flexible production plan, includes setting constraint conditions and setting optimization goals. Setting constraint conditions incorporates the device health status and fault prediction results into the production plan constraints to ensure that high-risk devices are given priority for maintenance tasks and low-risk devices undertake the main production tasks. Setting optimization goals refers to allocating tasks through an optimization algorithm, and the objective function is to maximize production efficiency.

6. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: The said step 5 is collaborative operation execution. After the optimized plan is issued to the workshop production control system, the equipment execution follows the following process: Equipment in good health status gives priority to executing high-priority tasks to maximize production capacity and resource utilization rate and ensure the timely completion of key production tasks; Equipment in a warning state undertakes short-cycle or low-risk tasks, avoids long-term high-load operation, and arranges for early maintenance to prevent the state from deteriorating further. The maintenance work and production tasks are carried out alternately to minimize the downtime; Equipment in poor health status stops running immediately, and maintenance and repair measures are implemented preferentially. After the maintenance is completed and the equipment returns to a healthy state, the equipment can be put back into production. During this process, the production plan is dynamically adjusted, and other equipment shares its tasks to ensure that the overall production progress is not significantly affected.

7. A collaborative operation method for equipment predictive maintenance and flexible production manufacturing equipment according to claim 1, characterized in that: The said step 6 is maintenance and production result evaluation. The equipment maintenance and production result evaluation includes three core indicators: equipment utilization rate, average maintenance time, and unplanned downtime rate. The equipment utilization rate measures the ratio of the actual running time of the equipment to the total time, reflecting the operating efficiency of the equipment. The average maintenance time calculates the average value of each maintenance time to help evaluate the timeliness and efficiency of the maintenance work. The unplanned downtime rate calculates the proportion of the unplanned downtime time in the total time, which is an important indicator for evaluating the system stability. Comprehensive analysis of these indicators helps to timely detect the weak links in the equipment operation, optimize the maintenance strategy, and improve the equipment reliability and production efficiency.

Images