A distributed real-time computing scheduling system and method

Abstract

The present invention relates to a distributed real-time computing scheduling system and method, belonging to the technical field of control or regulation systems for non-electric variables, and comprising the following steps: Step 1): Obtain the target data after converting distributed non-electric variables into respective electric signals; Step 2): Determine the real-time computing scheduling target according to the target data of each electric signal, and establish a scheduling model for the real-time computing scheduling target; Step 3): Establish a real-time computing scheduling strategy according to the scheduling target preference in the scheduling model; Step 4): Real-time monitor the scheduling model of the real-time computing scheduling target in Step 2) and the real-time computing scheduling strategy in Step 3); The beneficial effects of the present invention: According to the established scheduling model of the real-time computing scheduling target and the established real-time computing scheduling strategy, realize the real-time monitoring and scheduling of the scheduling model of the real-time computing scheduling target and the real-time computing scheduling strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of control or regulation systems for non - electrical variables, and particularly relates to a distributed real - time computing scheduling system and method. Background Art

[0002] Non - electrical variables are non - electrical quantities, such as temperature, pressure, speed, displacement, strain, flow rate, and liquid level, etc. The measurement of non - electrical variables usually requires sensors to convert non - electrical physical quantities into electrical signals, and then process and display them. For example, a temperature sensor can convert temperature changes into electrical signals, and a pressure sensor can convert pressure changes into electrical signals, and these electrical signals can be further used for control and monitoring.

[0003] Distributed non - electrical variables convert each decentralized non - electrical variable in the system into different electrical signals. After scheduling the electrical signals of different types of non - electrical variables, they can be closely linked together. After the distributed non - electrical variables are converted into their respective electrical signals, real - time scheduling is performed on each electrical signal. The real - time scheduling of each electrical signal has changed the traditional scheduling of different electrical signals and is developing towards a more flexible, sustainable, and intelligent direction. Real - time scheduling of this distributed non - electrical variable system can maximize the utilization and scheduling of non - electrical variables and improve the scheduling efficiency of distributed non - electrical variables. Summary of the Invention

[0004] The technical problem to be solved by the present invention is how to perform real - time scheduling on each electrical signal. The present invention provides a distributed real - time computing scheduling system and method, which find out a certain sub - data corresponding to each in each target data, classify the found sub - data, establish a scheduling model for real - time computing scheduling objectives according to the classified sub - data, and then establish a real - time computing scheduling strategy according to the scheduling objective preferences to realize the real - time monitoring and scheduling of the scheduling model and real - time computing scheduling strategy for real - time computing scheduling objectives.

[0005] To achieve the above object, the present invention is realized through the following technical solutions:

[0006] A distributed real - time computing scheduling method, comprising the following steps:

[0007] Step 1): Obtain the target data after converting distributed non - electrical variables into their respective electrical signals;

[0008] Step 2): Determine the real - time computing scheduling objective according to the target data of each electrical signal, and establish a scheduling model for the real - time computing scheduling objective; wherein, the scheduling model for the real - time computing scheduling objective is established by using the target vector method;

[0009] Step 3): Establish a real-time computing scheduling policy according to the scheduling target preferences in the scheduling model; among them, the real-time computing scheduling policy is established in a real-time preference manner;

[0010] Step 4): Real-time monitor the scheduling model of the real-time computing scheduling target in Step 2) and the real-time computing scheduling policy in Step 3).

[0011] Optionally, in Step 1), classify the target data of the respective electrical signals by the data grid method, and the data grid method is as follows in formula (1):

[0012] (1);

[0013] Wherein, , and are the first target data, the second target data, and the third target data respectively; , and are a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data respectively; is the target orientation operation; , and are that the first target data is target-oriented to a certain sub-data of the first target data, the second target data is target-oriented to a certain sub-data of the second target data, and the third target data is target-oriented to a certain sub-data of the third target data respectively; is the data connection operation, connects each sub-data; is a certain time period, is to adjust a certain time period;

[0014] is to extract a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data;

[0015] is to extract a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data per unit time;

[0016] is to classify a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data respectively; is each sub-data after classification.

[0017] Optionally, in Step 1), classify the target data of the respective electrical signals by the data grid method, and the following steps are adopted:

[0018] Step a): During a certain time period, each target data is respectively targeted at its corresponding sub-data among the respective target data;

[0019] Step b): Extract the corresponding sub-data in each target data within a unit time;

[0020] Step c): Classify each sub-data respectively.

[0021] Optionally, in step 2), the target vector method is as follows in formulas (2) - (3):

[0022] (2);

[0023] (3);

[0024] In formula (2), are the classified sub-data, is a selected sub-data after classification, is a triggering operation performed within a fixed unit time, is a fixed unit time, is to select a classified sub-data from the classified sub-data within a fixed unit time; is a scheduling model; is to send the selected and classified sub-data into the scheduling model inside the operation; is to send the selected and classified sub-data to the scheduling model inside; is to select a classified sub-data from the classified sub-data within a fixed unit time and send it into the scheduling model inside, is a scheduling model for real-time computing of scheduling objectives;

[0025] In formula (3), is a sub-data of the first target data, is a sub-data of the second target data, is a sub-data of the third target data, and the scheduling model is formed by the scheduling of each sub-data .

[0026] Optionally, in step 2), the target vector method is as follows:

[0027] Step A): Triggeringly select a classified sub-data within a fixed unit time;

[0028] Step B): Send a selected and classified sub-data into the scheduling model to construct a scheduling model for real-time computing scheduling objectives.

[0029] Optionally, in step 3), the real-time preference method is as follows formula (4):

[0030] (4);

[0031] Wherein, is a certain sub-data randomly selected from the first target data, is a certain sub-data randomly selected from the second target data; is a certain sub-data randomly selected from the third target data; is to connect the randomly selected sub-data; is a fixed unit of time; is the scheduling target preference; is the scheduling operation;

[0032] is to establish a real-time computing scheduling strategy according to the scheduling target preference in the scheduling model; is the real-time computing scheduling strategy.

[0033] Optionally, in step 3), the real-time preference method is as follows steps:

[0034] Step a'): Randomly select the sub-data corresponding to each of the target data in each target data within a fixed unit of time;

[0035] Step b'): Establish a real-time computing scheduling strategy according to the scheduling target preference in the scheduling model.

[0036] Optionally, in step 4), according to the real-time computing scheduling objective and the scheduling target preference, monitor the scheduling model and the real-time computing scheduling strategy of the real-time computing scheduling objective in real time.

[0037] A distributed real-time computing scheduling system, comprising:

[0038] A data classification module for classifying a certain sub-data corresponding to each of the target data;

[0039] A data scheduling module for constructing a scheduling model for real-time computing scheduling objectives, and the data scheduling module is connected to the data classification module;

[0040] A scheduling target preference module for establishing a real-time computing scheduling strategy, and the scheduling target preference module is connected to the data scheduling module;

[0041] A monitoring module is used to monitor in real time the scheduling model of the real-time computing scheduling target in the data scheduling module and the real-time computing scheduling policy in the data scheduling module. The monitoring module is connected to the data scheduling module and the scheduling target preference module;

[0042] Each target data is classified for a corresponding sub-data through a data classification module. After classification, each sub-data establishes a scheduling model of the real-time computing scheduling target in the data scheduling module and a real-time computing scheduling policy in the scheduling target preference module. Then, the monitoring module monitors in real time the scheduling model of the real-time computing scheduling target and the real-time computing scheduling policy.

[0043] Advantages of the present invention:

[0044] The present invention mainly finds out a corresponding sub-data for each target data, classifies the found sub-data, establishes a scheduling model of the real-time computing scheduling target according to the classified sub-data, and then establishes a real-time computing scheduling policy according to the scheduling target preference, so as to realize the real-time monitoring and scheduling of the scheduling model of the real-time computing scheduling target and the real-time computing scheduling policy. Description of the drawings

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0046] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0047] Figure 2 It is a working flowchart of the present invention. Detailed implementation manners

[0048] The following will describe in detail the embodiments of the present application with reference to the drawings.

[0049] The terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0050] In the description of the present application, it should be noted that unless otherwise clearly specified and defined, the terms "connected" and "connected to" should be understood in a broad sense. For example, it can be welding, bolt connection, riveting, or connection with data correlation; it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations. Embodiment

[0051] As Figure 1 shown, as Figure 1 shown, this embodiment provides a distributed real-time computing scheduling system, including:

[0052] The data acquisition module is used to obtain the target data after converting distributed non-electrical variables into respective electrical signals;

[0053] The data classification module is used for classifying a certain sub-data corresponding to each of the target data;

[0054] The data scheduling module is used to establish a scheduling model for real-time computing scheduling objectives, and the data scheduling module is connected to the data classification module;

[0055] The scheduling objective preference module is used to establish a real-time computing scheduling strategy, and the scheduling objective preference module is connected to the data scheduling module;

[0056] The monitoring module is used to monitor in real time the scheduling model of the real-time computing scheduling objective in the data scheduling module and the real-time computing scheduling strategy in the data scheduling module. The monitoring module is connected to the data scheduling module and the scheduling objective preference module;

[0057] The data acquisition module obtains the target data after converting distributed non-electrical variables into respective electrical signals, classifies a certain sub-data corresponding to each of the target data through the data classification module, establishes a scheduling model for real-time computing scheduling objectives in the data scheduling module and a real-time computing scheduling strategy in the scheduling objective preference module for the classified sub-data, and then the monitoring module monitors in real time the scheduling model of the real-time computing scheduling objective and the real-time computing scheduling strategy. Embodiment

[0058] Based on Embodiment 1, as Figure 2 shown, this embodiment provides a distributed real-time computing scheduling method, including the following steps:

[0059] Step 1): Obtain the target data after converting distributed (pressure, displacement or strain) non-electrical variables into respective electrical signals;

[0060] Step 2): Determine the real-time calculation scheduling objective according to the target data of each electrical signal, and establish a scheduling model for the real-time calculation scheduling objective; among them, the scheduling model for the real-time calculation scheduling objective is established by using the target vector method;

[0061] Step 3): Establish a real-time calculation scheduling strategy according to the scheduling objective preference in the scheduling model; among them, the real-time calculation scheduling strategy is established by using the real-time preference method;

[0062] Step 4): Monitor the scheduling model in real time.

[0063] Specifically, in step 1), the target data of each electrical signal is classified by using the data grid method, and the data grid method is as follows formula (1):

[0064] (1);

[0065] Among them, , and are the first target data, the second target data and the third target data respectively; , and are a certain sub-data of the first target data, a certain sub-data of the second target data and a certain sub-data of the third target data respectively (there are multiple sub-data in the first target data , multiple sub-data in the second target data and multiple sub-data in the third target data . Select a certain sub-data in the first target data , select a certain sub-data in the second target data , and select a certain sub-data in the third target data ); is the target orientation operation; , and are that the first target data is targeted at a certain sub-data of the first target data, the second target data is targeted at a certain sub-data of the second target data, and the third target data is targeted at a certain sub-data of the third target data respectively; is the data connection operation, connects each sub-data; is a certain time period, is to adjust a certain time period (that is, the length of a certain time period can be adjusted);

[0066] Here it needs to be explained that the target orientation operation is performed according to the set situation, that is, the first target data Target orientation operation According to the setting of the target orientation operation, target orientation is performed on a certain sub - data of the first target data; similarly, for the second target data Target orientation operation According to the setting of the target orientation operation, target orientation is performed on a certain sub - data of the second target data; for the third target data Target orientation operation According to the setting of the target orientation operation, target orientation is performed on a certain sub - data of the third target data.

[0067] It is to extract a certain sub - data of the first target data, a certain sub - data of the second target data, and a certain sub - data of the third target data;

[0068] It is to extract a certain sub - data of the first target data, a certain sub - data of the second target data, and a certain sub - data of the third target data per unit time;

[0069] It is to classify each of the certain sub - data of the first target data, the second target data, and the third target data respectively; They are the classified sub - data.

[0070] In addition, in step 1), the target data of the respective electrical signals is classified using the data grid method, and the following steps are adopted:

[0071] Step a): During a certain time period, each target data is target - oriented to its corresponding sub - data in each target data;

[0072] Step b): Extract the corresponding sub - data in each target data per unit time;

[0073] Step c): Classify each sub - data respectively.

[0074] In step 2), the target vector method is as the following formulas (2) - (3):

[0075] (2);

[0076] (3);

[0077] In formula (2), They are the classified sub - data, is to select a certain sub - data after classification, is the trigger operation performed within a fixed unit time (i.e., the trigger operation is performed every few seconds), is the fixed unit time, To select a certain classified sub - data from each of the classified sub - data within a fixed unit of time; It is a scheduling model; It is the operation of sending a certain selected and classified sub - data into the scheduling model inside; It is to send a certain selected and classified sub - data to the scheduling model inside; It is to send, within a fixed unit of time, a certain classified sub - data selected from each of the classified sub - data to the scheduling model inside, It is a scheduling model for real - time calculating scheduling objectives;

[0078] In formula (3), is a certain sub - data of the first target data, is a certain sub - data of the second target data, is a certain sub - data of the third target data, and the scheduling of each sub - data forms a scheduling model .

[0079] Combining formula (2) and formula (3), a certain classified sub - data in the first target data , a certain classified sub - data in the second target data , a certain classified sub - data in the third target data are sent into the scheduling model inside, and finally a scheduling model for real - time calculating scheduling objectives is constructed .

[0080] In addition, in step 2), the target vector method is as follows:

[0081] Step A): Triggeringly select a certain classified sub - data within a fixed unit of time;

[0082] Step B): Send the selected and classified sub - data into the scheduling model to construct a scheduling model for real - time calculating scheduling objectives.

[0083] In step 3), the real - time preference method is as follows in formula (4):

[0084] (4);

[0085] Among them, is a certain randomly selected sub - data in the first target data, is a certain randomly selected sub - data in the second target data; is a certain randomly selected sub - data in the third target data; To concatenate each randomly selected sub - data; is a fixed unit time; is the scheduling target preference ( biased towards a certain sub - data randomly selected from the first target data , or a certain sub - data randomly selected from the second target data , or a certain sub - data randomly selected from the third target data ); is the scheduling operation;

[0086] is to establish a real - time computing scheduling strategy according to the scheduling target preference in the scheduling model; is the real - time computing scheduling strategy.

[0087] Formula (4) randomly selects sub - data in real time according to the scheduling target preference or sub - data or sub - data .

[0088] In addition, in step 3), the real - time preference method is as follows:

[0089] Step a'): Randomly select the respective corresponding sub - data of each of the target data within a fixed unit time;

[0090] Step b'): Establish a real - time computing scheduling strategy according to the scheduling target preference in the scheduling model.

[0091] In step 4), according to the real - time computing scheduling target and the scheduling target preference, real - time monitor the scheduling model and the real - time computing scheduling strategy of the real - time computing scheduling target. Embodiment

[0092] Based on Embodiment 2, the present invention mainly finds out a certain sub - data corresponding to each of the target data, classifies the found sub - data, establishes a scheduling model for the real - time computing scheduling target according to the classified sub - data, and then establishes a real - time computing scheduling strategy according to the scheduling target preference. Embodiment

[0093] Based on Embodiment 2, in formula (1), the first target data has multiple sub - data. According to the setting of the target - directed operation , target - direct a certain sub - data in the first target data ; Similarly, the second target data has multiple sub - data. According to the setting of the target - directed operation , target - direct a certain sub - data in the second target data ; The third target data has multiple sub-data, and according to the setting of the target orientation operation , target-orient the third target data to a certain sub-data .

[0094] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope recorded in the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims described above.

Claims

1. A distributed real-time computing scheduling method, characterized in that It includes the following steps: Step 1): Obtain the target data after converting distributed non - electrical variables into respective electrical signals; In the said Step 1), classify the target data of the respective electrical signals by using the data grid method, and the data grid method is as formula (1) below: (1)； Among them, , and are the first target data, the second target data, and the third target data respectively; , and are a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data respectively; is the target orientation operation; , and are that the first target data is target-oriented to a certain sub-data of the first target data, the second target data is target-oriented to a certain sub-data of the second target data, and the third target data is target-oriented to a certain sub-data of the third target data respectively; is the data connection operation, connects each sub-data; is a certain time period, is to adjust a certain time period; To extract a certain sub-data of the first target data, a certain sub-data of the second target data, and a certain sub-data of the third target data; Extract a certain sub - data of the first target data, a certain sub - data of the second target data, and a certain sub - data of the third target data per unit time; To classify a certain sub - data of the first target data, a certain sub - data of the second target data, and a certain sub - data of the third target data respectively; For each of the classified sub - data; Step 2): Determine the real - time calculation scheduling target according to the target data of each electrical signal respectively, and establish a scheduling model for the real - time calculation scheduling target; among them, establish a scheduling model for the real - time calculation scheduling target by using the target vector method; In the said Step 2), the target vector method is as formula (2) - formula (3) below: (2)； (3)； In formula (2), are the classified sub - data,[ is a selected sub - data after classification,[ is a triggering operation performed within a fixed unit time,[ is the fixed unit time,[ is to select a classified sub - data from the classified sub - data within a fixed unit time;[ is the scheduling model;[ is the operation of sending the selected and classified sub - data into the scheduling model[ inside;[ is to send the selected and classified sub - data to the scheduling model[ inside;[ is to send the selected and classified sub - data from the classified sub - data into the scheduling model[ inside,[ is the scheduling model for real - time calculating the scheduling target[ In formula (3), is a certain sub - data of the first target data, is a certain sub - data of the second target data, is a certain sub - data of the third target data. A scheduling model is formed through the scheduling of each sub - data ; Step 3): Establish a real - time calculation scheduling strategy according to the scheduling target preference in the scheduling model; among them, establish a real - time calculation scheduling strategy by using the real - time preference method; In the said Step 3), the real - time preference method is as formula (4) below: (4)； wherein, is a certain sub - data randomly selected from the first target data, is a certain sub - data randomly selected from the second target data; is a certain sub - data randomly selected from the third target data; is to concatenate the randomly selected sub - data; is a fixed unit of time; is the scheduling target preference; is the scheduling operation; To establish a real-time computing scheduling strategy according to the scheduling objective preferences in the scheduling model; For the real-time computing scheduling strategy; Step 4): Real - time monitor the scheduling model of the real - time calculation scheduling target in Step 2) and the real - time calculation scheduling strategy in Step 3).

2. The distributed real-time computing scheduling method according to claim 1, wherein In the said Step 1), classify the target data of the respective electrical signals by using the data grid method, and adopt the following steps: Step a): In a certain time period, direct each target data to its corresponding sub - data in each target data; Step b): Extract the corresponding sub - data in each target data within a unit time; Step c): Classify each sub - data respectively.

3. A distributed real-time computing scheduling method according to claim 1, characterized in that In the said Step 2), the target vector method is as follows: Step A): Triggeringly select a certain classified sub - data within a fixed unit time; Step B): Send the selected and classified certain sub - data into the scheduling model to construct a scheduling model for the real - time calculation scheduling target.

4. A distributed real-time computing scheduling method according to claim 1, characterized in that, In the said Step 3), the real - time preference method is as follows: Step a'): Randomly select the corresponding sub - data of each target data in each target data within a fixed unit time; Step b'): Establish a real - time calculation scheduling strategy according to the scheduling target preference in the scheduling model.

5. A distributed real-time computing scheduling method according to claim 1, characterized in that In the said Step 4), according to the real - time calculation scheduling target and the scheduling target preference, real - time monitor the scheduling model of the real - time calculation scheduling target and the real - time calculation scheduling strategy.

6. A distributed real-time computing scheduling system for executing a distributed real-time computing scheduling method according to any one of claims 1-5, characterized in that, It includes: A data classification module, used for classifying a certain sub - data corresponding to each of the target data; A data scheduling module, used for establishing a scheduling model for the real - time calculation scheduling target, and the data scheduling module is connected to the data classification module; A scheduling target preference module, used for establishing a real - time calculation scheduling strategy, and the scheduling target preference module is connected to the data scheduling module; A monitoring module, used for real - time monitoring the scheduling model of the real - time calculation scheduling target in the data scheduling module and the real - time calculation scheduling strategy in the data scheduling module, and the monitoring module is connected to the data scheduling module and the scheduling target preference module; Classify a certain sub - data corresponding to each target data through the data classification module. After classification, establish a scheduling model for real - time computing scheduling targets for each sub - data in the data scheduling module and establish a real - time computing scheduling policy in the scheduling target preference module. Then, the monitoring module monitors the scheduling model of the real - time computing scheduling target and the real - time computing scheduling policy in real time.

Images