# US 2026/0312820 A1 · Agent-based production modeling with deterministic and probabilistic verification gating

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0312820 A1
- Document ID: US20260312820A1
- Status: Published application
- Kind: A1
- Application number: 19/577,473
- Filed: 2026-03-25
- Published: 2026-10-08
- Earliest claimed: 2025-04-07
- Assignees: EnergySolutions Al LLC
- Applicants: EnergySolutions Al LLC
- Inventors: Michael Dean Orr, JR.
- Web page: https://patentgrep.com/patent/US20260312820A1

## CPC

- H04L 9/3236: {using cryptographic hash functions}
- H04L 9/50: {using hash chains, e.g. blockchains or hash trees}
- H04L 2209/56: Financial cryptography, e.g. electronic payment or e-cash
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Apparatus and method for production reconciliation in a low trust environment, such as but not limited to a multistream hydrocarbon production system. A reconciliation run is executed by a data plane of a processor-based system to generate candidate outputs representing reconciled production values, and to assemble a corresponding evidence bundle keyed by a run_id. A truth gate module evaluates the candidate outputs using at least one deterministic validation criterion and at least one probabilistic validation criterion to generate a composite truth score. Promotion of the candidate outputs to a validated semantic state results in posting to a system of record for settlement, such as a private or public distributed ledger. Promotion is blocked either due to failure of the deterministic validation criterion or the composite truth score falling below the predetermined threshold. Exception events are handled via human intervention using an alert register of a user plane.

## Claims

### Claim 1

A computer-implemented method for validating a reconciliation run in an agent-based production reconciliation system, comprising:
executing, by a data plane of the system, a reconciliation run associated with a unique run identifier (run_id) to generate candidate outputs representing reconciled production values derived from accumulated measurements of a physical production system;
assembling, for the reconciliation run, an evidence bundle keyed to the run_id and comprising raw measurements, intermediate calculations, agent validation records, and a cryptographic hash of input data;
evaluating, by a truth gate executed by a programmable processor, the candidate outputs using at least one deterministic validation criterion that produces a non-overridable pass/fail result based on schema-level conditions of the accumulated measurements, and at least one probabilistic validation criterion that generates a numerical score representing a likelihood that the candidate outputs are trustworthy, the probabilistic validation criterion including at least a reconciliation-history factor derived from prior reconciliation runs associated with the run_id and an evidence-bundle integrity factor derived from comparing the cryptographic hash to a stored hash of the evidence bundle;
computing, by the truth gate, a composite truth score for the reconciliation run based on the probabilistic validation criterion; and
promoting the candidate outputs to a validated semantic state responsive to the deterministic validation criterion passing and responsive to the composite truth score exceeding a predetermined threshold, and blocking the candidate outputs from promotion responsive to either the deterministic validation criterion failing or the composite truth score falling below the predetermined threshold, wherein promotion to the validated semantic state authorizes posting of the candidate outputs to a system of record for settlement.

### Claim 2 (depends on 1)

The method of claim 1, wherein the deterministic validation criterion comprises verifying that all required measurement nodes for the reconciliation run are present in the evidence bundle.

### Claim 3 (depends on 1)

The method of claim 1, wherein the deterministic validation criterion comprises verifying that a sum of degrees-of-interest (DOI) for a product stream equals exactly one (1) up to a predetermined plurality of decimal places of precision.

### Claim 4 (depends on 3)

The method of claim 3, wherein the predetermined plurality of decimal places of precision is eight (8) decimal places so that the deterministic validation criterion results in a pass for a value of 1.0000000±0.0000001 and results in a fail for a sum of 0.9999999 or lower.

### Claim 5 (depends on 1)

The method of claim 1, wherein the deterministic validation criterion comprises verifying that a division-order ownership table is complete and internally consistent so that all owners of interest have legally certified entitlement claims.

### Claim 6 (depends on 1)

The method of claim 1, wherein the deterministic validation criterion comprises verifying that no curative block is active for the reconciliation run.

### Claim 7 (depends on 1)

The method of claim 1, wherein the probabilistic validation criterion comprises a completeness score based on a percentage of expected measurements present in the evidence bundle.

### Claim 8 (depends on 1)

The method of claim 1, wherein the probabilistic validation criterion comprises a freshness score based on a recency of the measurements relative to a timestamp of the reconciliation run.

### Claim 9 (depends on 1)

The method of claim 1, wherein the probabilistic validation criterion comprises an agent-confidence score derived from model-generated confidence values associated with intermediate calculations.

### Claim 10 (depends on 1)

The method of claim 1, wherein the probabilistic validation criterion comprises a reconciliation-history score based on a weighted average of truth scores from prior reconciliation runs associated with the run_id.

### Claim 11 (depends on 1)

The method of claim 1, wherein computing the composite truth score comprises applying a weighted combination of the probabilistic validation criteria.

### Claim 12 (depends on 1)

The method of claim 1, further comprising writing data associated with the candidate outputs having the validated semantic state to a blockchain-based distributed ledger in a local memory of the system, and writing data associated with the evidence bundle to a cold storage data storage device.

### Claim 13 (depends on 1)

The method of claim 1, further comprising generating and updating a Run Registry as a table structure in memory that records the run_id and data associated with each reconciliation run process, wherein the Run Registry is updated with a complete indication responsive to the associated candidate output receiving the validated semantic state indication from the truth gate.

### Claim 14 (depends on 1)

The method of claim 1, further comprising utilizing an alert register of a user plane to process, via human input, a human-workflow exception event resulting from failure of at least one of the deterministic validation criterion or the probabilistic validation criterion of the truth gate.

### Claim 15 (depends on 1)

The method of claim 1, wherein the reconciliation run is executed in connection with an oil and gas production system that generates a plurality of hydrocarbon and water streams comprising natural gas, natural gas liquids (NGL), oil, condensate, and produced water, and wherein the evaluating step applies the deterministic validation criterion and the probabilistic validation criterion to candidate outputs associated with at least one of the natural gas, NGL, oil, condensate, or produced water streams.

### Claim 16 (depends on 1)

The method of claim 1, further comprising accessing a digital twin of the physical production system on a display of a network accessible device in a form of a web-based software application, the web-based software application displaying real time state of the physical production system including results generated by the truth gate module.

### Claim 17

A system for validating a reconciliation run in an agent-based production reconciliation environment, comprising one or more processors and memory storing instructions that, when executed, cause the system to:
maintain a run registry that stores a unique run identifier (run_id) for each reconciliation run;
generate, for each reconciliation run, an evidence bundle keyed to the run_id and comprising raw measurements, intermediate calculations, agent-generated candidate outputs, validation artifacts, and a cryptographic hash of input data; and
execute a truth gate module configured to evaluate the candidate outputs using at least one deterministic validation criterion that produces a non-overridable pass/fail result based on schema-level conditions of the accumulated measurements, and at least one probabilistic validation criterion that generates a numerical score representing a likelihood that the candidate outputs are trustworthy;
wherein the probabilistic validation criterion includes at least a reconciliation-history factor derived from prior reconciliation runs associated with the run_id and an evidence-bundle integrity factor derived from comparing a hash of current inputs to a stored hash of the evidence bundle;
wherein the truth gate module computes a composite truth score for the reconciliation run based on the probabilistic validation criterion, promotes the candidate outputs to a validated semantic state responsive to the deterministic validation criterion passing and the composite truth score exceeding a predetermined threshold, and blocks the candidate outputs from promotion responsive to either the deterministic validation criterion failing or the composite truth score falling below the predetermined threshold; and
wherein promotion to the validated semantic state authorizes posting of the candidate outputs to a system of record for settlement comprising a blockchain-based distributed ledger.

### Claim 18 (depends on 17)

The system of claim 17, wherein the deterministic validation criterion applied by the truth gate module is a pass/fail hard encoded test comprising each of the following: a verification that all required measurement nodes for the reconciliation run are present in the evidence bundle, a verification that a sum of Divisions of Interest (DOI) for a product stream equals exactly one (1) up to a predetermined plurality of decimal places of precision, and that a division-order ownership table is complete and internally consistent so that all owners of interest have legally certified entitlement claims.

### Claim 19 (depends on 17)

The system of claim 17, wherein the probabilistic validation criterion applied by the truth gate module is composite numerical truth score comprising each of the following components: a completeness score based on a percentage of expected measurements present in the evidence bundle, a freshness score based on a recency of the measurements relative to a timestamp of the reconciliation run, an agent-confidence score derived from model-generated confidence values associated with intermediate calculations, and a reconciliation-history score based on a weighted average of truth scores from prior reconciliation runs associated with the run_id.

### Claim 20 (depends on 17)

The system of claim 17, wherein the reconciliation run is executed in connection with an oil and gas production system that generates a plurality of hydrocarbon and water streams comprising natural gas, natural gas liquids (NGL), oil, condensate, and produced water, and wherein the truth gate module applies the deterministic validation criterion and the probabilistic validation criterion to candidate outputs associated with at least one of the natural gas, NGL, oil, condensate, or produced water streams.
